arxiv_id
string | title
string | authors
list | submission_date
string | comments
string | primary_subject
string | subjects
string | doi
string | abstract
string | file_path
string |
|---|---|---|---|---|---|---|---|---|---|
2212.09126
|
Pigeonhole Stochastic Gradient Langevin Dynamics for Large Crossed Mixed Effects Models
|
[
"Xinyu Zhang",
"Cheng Li"
] |
18 Dec 2022 (<a href="https://arxiv.org/abs/2212.09126v1">v1</a>), last revised 9 May 2025 (this version, v2)
|
Computation (stat.CO)
|
Computation (stat.CO)
|
https://doi.org/10.48550/arXiv.2212.09126
|
Large crossed mixed effects models with imbalanced structures and missing data pose major computational challenges for standard Bayesian posterior sampling algorithms, as the computational complexity is usually superlinear in the number of observations. We propose two efficient subset-based stochastic gradient MCMC algorithms for such crossed mixed effects models, which facilitate scalable inference on both the variance components and regression coefficients. The first algorithm is developed for balanced design without missing observations, where we leverage the closed-form expression of the precision matrix for the full data matrix. The second algorithm, which we call the pigeonhole stochastic gradient Langevin dynamics (PSGLD), is developed for both balanced and unbalanced designs with potentially a large proportion of missing observations. Our PSGLD algorithm imputes the latent crossed random effects by running short Markov chains and then samples the model parameters of variance components and regression coefficients at each MCMC iteration. We provide theoretical guarantees by showing the convergence of the output distribution from the proposed algorithms to the target non-log-concave posterior distribution. A variety of numerical experiments based on both synthetic and real data demonstrate that the proposed algorithms can significantly reduce the computational cost of the standard MCMC algorithms and better balance the approximation accuracy and computational efficiency.
|
stat-05.zip/2212.09126.pdf
|
|
2212.08549
|
Microcanonical Hamiltonian Monte Carlo
|
[
"Jakob Robnik",
"G. Bruno De Luca",
"Eva Silverstein",
"Uroš Seljak"
] |
16 Dec 2022 (<a href="https://arxiv.org/abs/2212.08549v1">v1</a>), last revised 22 Aug 2023 (this version, v2)
|
34 pages, 11 figures
|
Computation (stat.CO)
|
Computation (stat.CO); Instrumentation and Methods for Astrophysics (astro-ph.IM); High Energy Physics - Lattice (hep-lat); High Energy Physics - Theory (hep-th)
|
https://doi.org/10.48550/arXiv.2212.08549
|
We develop Microcanonical Hamiltonian Monte Carlo (MCHMC), a class of models which follow a fixed energy Hamiltonian dynamics, in contrast to Hamiltonian Monte Carlo (HMC), which follows canonical distribution with different energy levels. MCHMC tunes the Hamiltonian function such that the marginal of the uniform distribution on the constant-energy-surface over the momentum variables gives the desired target distribution. We show that MCHMC requires occasional energy conserving billiard-like momentum bounces for ergodicity, analogous to momentum resampling in HMC. We generalize the concept of bounces to a continuous version with partial direction preserving bounces at every step, which gives an energy conserving underdamped Langevin-like dynamics with non-Gaussian noise (MCLMC). MCHMC and MCLMC exhibit favorable scalings with condition number and dimensionality. We develop an efficient hyperparameter tuning scheme that achieves high performance and consistently outperforms NUTS HMC on several standard benchmark problems, in some cases by more than an order of magnitude.
|
stat-05.zip/2212.08549.pdf
|
2308.09060
|
TraitLab: a Matlab package for fitting and simulating binary tree-like data
|
[
"Luke J. Kelly",
"Geoff K. Nicholls",
"Robin J. Ryder",
"David Welch"
] |
17 Aug 2023
|
Manual describing the TraitLab software for phylogenetic inference
|
Computation (stat.CO)
|
Computation (stat.CO)
|
https://doi.org/10.48550/arXiv.2308.09060
|
TraitLab is a software package for simulating, fitting and analysing tree-like binary data under a stochastic Dollo model of evolution. The model also allows for rate heterogeneity through catastrophes, evolutionary events where many traits are simultaneously lost while new ones arise, and borrowing, whereby traits transfer laterally between species as well as through ancestral relationships. The core of the package is a Markov chain Monte Carlo (MCMC) sampling algorithm that enables the user to sample from the Bayesian joint posterior distribution for tree topologies, clade and root ages, and the trait loss, catastrophe and borrowing rates for a given data set. Data can be simulated according to the fitted Dollo model or according to a number of generalized models that allow for heterogeneity in the trait loss rate, biases in the data collection process and borrowing of traits between lineages. Coupled pairs of Markov chains can be used to diagnose MCMC mixing and convergence and to debias MCMC estimators. The raw data, MCMC run output, and model fit can be inspected using a number of useful graphical and analytical tools provided within the package or imported into other popular analysis programs. TraitLab is freely available and runs within the Matlab computing environment with its Statistics and Machine Learning toolbox, no other additional toolboxes are required.
|
stat-05.zip/2308.09060.pdf
|
2212.05555
|
Bayesian inversion with α-stable priors
|
[
"Jarkko Suuronen",
"Tomás Soto",
"Neil K. Chada",
"Lassi Roininen"
] |
11 Dec 2022 (<a href="https://arxiv.org/abs/2212.05555v1">v1</a>), last revised 23 Jun 2023 (this version, v2)
|
Computation (stat.CO)
|
Computation (stat.CO)
|
https://doi.org/10.48550/arXiv.2212.05555
|
We propose to use Lévy {\alpha}-stable distributions for constructing priors for Bayesian inverse problems. The construction is based on Markov fields with stable-distributed increments. Special cases include the Cauchy and Gaussian distributions, with stability indices {\alpha} = 1, and {\alpha} = 2, respectively. Our target is to show that these priors provide a rich class of priors for modelling rough features. The main technical issue is that the {\alpha}-stable probability density functions do not have closed-form expressions in general, and this limits their applicability. For practical purposes, we need to approximate probability density functions through numerical integration or series expansions. Current available approximation methods are either too time-consuming or do not function within the range of stability and radius arguments needed in Bayesian inversion. To address the issue, we propose a new hybrid approximation method for symmetric univariate and bivariate {\alpha}-stable distributions, which is both fast to evaluate, and accurate enough from a practical viewpoint. Then we use approximation method in the numerical implementation of {\alpha}-stable random field priors. We demonstrate the applicability of the constructed priors on selected Bayesian inverse problems which include the deconvolution problem, and the inversion of a function governed by an elliptic partial differential equation. We also demonstrate hierarchical {\alpha}-stable priors in the one-dimensional deconvolution problem. We employ maximum-a-posterior-based estimation at all the numerical examples. To that end, we exploit the limited-memory BFGS and its bounded variant for the estimator.
|
stat-05.zip/2212.05555.pdf
|
|
2308.15613
|
Mixed Variational Flows for Discrete Variables
|
[
"Gian Carlo Diluvi",
"Benjamin Bloem-Reddy",
"Trevor Campbell"
] |
29 Aug 2023 (<a href="https://arxiv.org/abs/2308.15613v1">v1</a>), last revised 26 Feb 2024 (this version, v3)
|
Computation (stat.CO)
|
Computation (stat.CO); Machine Learning (cs.LG); Machine Learning (stat.ML)
|
https://doi.org/10.48550/arXiv.2308.15613
|
Variational flows allow practitioners to learn complex continuous distributions, but approximating discrete distributions remains a challenge. Current methodologies typically embed the discrete target in a continuous space - usually via continuous relaxation or dequantization - and then apply a continuous flow. These approaches involve a surrogate target that may not capture the original discrete target, might have biased or unstable gradients, and can create a difficult optimization problem. In this work, we develop a variational flow family for discrete distributions without any continuous embedding. First, we develop a measure-preserving and discrete (MAD) invertible map that leaves the discrete target invariant, and then create a mixed variational flow (MAD Mix) based on that map. Our family provides access to i.i.d. sampling and density evaluation with virtually no tuning effort. We also develop an extension to MAD Mix that handles joint discrete and continuous models. Our experiments suggest that MAD Mix produces more reliable approximations than continuous-embedding flows while being significantly faster to train.
|
stat-05.zip/2308.15613.pdf
|
|
2212.04400
|
The Lifebelt Particle Filter for robust estimation from low-valued count data
|
[
"Alice Corbella",
"Trevelyan J. McKinley",
"Paul J. Birrell",
"Daniela De Angelis",
"Anne M. Presanis",
"Gareth O. Roberts",
"Simon E. F. Spencer"
] |
8 Dec 2022 (<a href="https://arxiv.org/abs/2212.04400v1">v1</a>), last revised 4 Dec 2024 (this version, v3)
|
Computation (stat.CO)
|
Computation (stat.CO)
|
https://doi.org/10.48550/arXiv.2212.04400
|
Particle filtering methods can be applied to estimation problems in discrete spaces on bounded domains, to sample from and marginalise over unknown hidden states. As in continuous settings, problems such as particle degradation can arise: proposed particles can be incompatible with the data, lying in low probability regions or outside the boundary constraints, and the discrete system could result in all particles having weights of zero. In this paper we introduce the Lifebelt Particle Filter (LBPF), a novel method for robust likelihood estimation in low-valued count problems. The LBPF combines a standard particle filter with one (or more) lifebelt particles which, by construction, lie within the boundaries of the discrete random variables, and therefore are compatible with the data. A mixture of resampled and non-resampled particles allows for the preservation of the lifebelt particle, which, together with the remaining particle swarm, provides samples from the filtering distribution, and can be used to generate unbiased estimates of the likelihood. The main benefit of the LBPF is that only one or few, wisely chosen, particles are sufficient to prevent particle collapse. Differently from other methods, there is no need to increase the number of particles, and therefore the computational effort, in regions of the parameter space that generate less likely hidden states. The LBPF can be used within a pseudo-marginal scheme to draw inferences on static parameters, $ \boldsymbol{\theta} $, governing the system. We address here the estimation of a parameter governing probabilities of deaths and recoveries of hospitalised patients during an epidemic.
|
stat-05.zip/2212.04400.pdf
|
|
2301.07838
|
Robust Chauvenet Rejection: Powerful, but Easy to Use Outlier Detection for Heavily Contaminated Data Sets
|
[
"Nicholas Konz",
"Daniel E. Reichart"
] |
19 Jan 2023
|
Computation (stat.CO)
|
Computation (stat.CO); Instrumentation and Methods for Astrophysics (astro-ph.IM)
|
https://doi.org/10.48550/arXiv.2301.07838
|
In Maples et al. (2018) we introduced Robust Chauvenet Outlier Rejection, or RCR, a novel outlier rejection technique that evolves Chauvenet's Criterion by sequentially applying different measures of central tendency and empirically determining the rejective sigma value. RCR is especially powerful for cleaning heavily-contaminated samples, and unlike other methods such as sigma clipping, it manages to be both accurate and precise when characterizing the underlying uncontaminated distributions of data sets, by using decreasingly robust but increasingly precise statistics in sequence. For this work, we present RCR from a software standpoint, newly implemented as a Python package while maintaining the speed of the C++ original. RCR has been well-tested, calibrated and simulated, and it can be used for both one-dimensional outlier rejection and $n$-dimensional model-fitting, with or without weighted data. RCR is free to use for academic and non-commercial purposes, and the code, documentation and accompanying web calculator can be found and easily used online at <a href="https://github.com/nickk124/RCR" rel="external noopener nofollow" class="link-external link-https">this https URL</a>
|
stat-05.zip/2301.07838.pdf
|
|
2309.13557
|
Bayesian Parameter Inference for Partially Observed Diffusions using Multilevel Stochastic Runge-Kutta Methods
|
[
"Pierre Del Moral",
"Shulan Hu",
"Ajay Jasra",
"Hamza Ruzayqat",
"Xinyu Wang"
] |
24 Sep 2023
|
Computation (stat.CO)
|
Computation (stat.CO); Numerical Analysis (math.NA)
|
https://doi.org/10.48550/arXiv.2309.13557
|
We consider the problem of Bayesian estimation of static parameters associated to a partially and discretely observed diffusion process. We assume that the exact transition dynamics of the diffusion process are unavailable, even up-to an unbiased estimator and that one must time-discretize the diffusion process. In such scenarios it has been shown how one can introduce the multilevel Monte Carlo method to reduce the cost to compute posterior expected values of the parameters for a pre-specified mean square error (MSE). These afore-mentioned methods rely on upon the Euler-Maruyama discretization scheme which is well-known in numerical analysis to have slow convergence properties. We adapt stochastic Runge-Kutta (SRK) methods for Bayesian parameter estimation of static parameters for diffusions. This can be implemented in high-dimensions of the diffusion and seemingly under-appreciated in the uncertainty quantification and statistics fields. For a class of diffusions and SRK methods, we consider the estimation of the posterior expectation of the parameters. We prove that to achieve a MSE of $\mathcal{O}(\epsilon^2)$, for $\epsilon>0$ given, the associated work is $\mathcal{O}(\epsilon^{-2})$. Whilst the latter is achievable for the Milstein scheme, this method is often not applicable for diffusions in dimension larger than two. We also illustrate our methodology in several numerical examples.
|
stat-05.zip/2309.13557.pdf
|
|
2212.01865
|
Variational Inference for Semiparametric Bayesian Novelty Detection in Large Datasets
|
[
"Luca Benedetti",
"Eric Boniardi",
"Leonardo Chiani",
"Jacopo Ghirri",
"Marta Mastropietro",
"Andrea Cappozzo",
"Francesco Denti"
] |
4 Dec 2022
|
Computation (stat.CO)
|
Computation (stat.CO); Applications (stat.AP)
|
https://doi.org/10.48550/arXiv.2212.01865
|
After being trained on a fully-labeled training set, where the observations are grouped into a certain number of known classes, novelty detection methods aim to classify the instances of an unlabeled test set while allowing for the presence of previously unseen classes. These models are valuable in many areas, ranging from social network and food adulteration analyses to biology, where an evolving population may be present. In this paper, we focus on a two-stage Bayesian semiparametric novelty detector, also known as Brand, recently introduced in the literature. Leveraging on a model-based mixture representation, Brand allows clustering the test observations into known training terms or a single novelty term. Furthermore, the novelty term is modeled with a Dirichlet Process mixture model to flexibly capture any departure from the known patterns. Brand was originally estimated using MCMC schemes, which are prohibitively costly when applied to high-dimensional data. To scale up Brand applicability to large datasets, we propose to resort to a variational Bayes approach, providing an efficient algorithm for posterior approximation. We demonstrate a significant gain in efficiency and excellent classification performance with thorough simulation studies. Finally, to showcase its applicability, we perform a novelty detection analysis using the openly-available Statlog dataset, a large collection of satellite imaging spectra, to search for novel soil types.
|
stat-05.zip/2212.01865.pdf
|
|
2308.09769
|
Pigeons.jl: Distributed Sampling From Intractable Distributions
|
[
"Nikola Surjanovic",
"Miguel Biron-Lattes",
"Paul Tiede",
"Saifuddin Syed",
"Trevor Campbell",
"Alexandre Bouchard-Côté"
] |
18 Aug 2023
|
Computation (stat.CO)
|
Computation (stat.CO); Distributed, Parallel, and Cluster Computing (cs.DC)
|
https://doi.org/10.48550/arXiv.2308.09769
|
We introduce a software package, <a href="http://Pigeons.jl" rel="external noopener nofollow" class="link-external link-http">this http URL</a>, that provides a way to leverage distributed computation to obtain samples from complicated probability distributions, such as multimodal posteriors arising in Bayesian inference and high-dimensional distributions in statistical mechanics. <a href="http://Pigeons.jl" rel="external noopener nofollow" class="link-external link-http">this http URL</a> provides simple APIs to perform such computations single-threaded, multi-threaded, and/or distributed over thousands of MPI-communicating machines. In addition, <a href="http://Pigeons.jl" rel="external noopener nofollow" class="link-external link-http">this http URL</a> guarantees a property that we call strong parallelism invariance: the output for a given seed is identical irrespective of the number of threads and processes, which is crucial for scientific reproducibility and software validation. We describe the key features of <a href="http://Pigeons.jl" rel="external noopener nofollow" class="link-external link-http">this http URL</a> and the approach taken to implement a distributed and randomized algorithm that satisfies strong parallelism invariance.
|
stat-05.zip/2308.09769.pdf
|
|
2212.01317
|
Fast gap-filling of massive data by local-equilibrium conditional simulations on GPU
|
[
"M. Lach",
"M. Žukovič"
] |
2 Dec 2022
|
26 pages, 14 figures
|
Computation (stat.CO)
|
Computation (stat.CO); Statistical Mechanics (cond-mat.stat-mech)
|
https://doi.org/10.48550/arXiv.2212.01317
|
The ever-growing size of modern space-time data sets, such as those collected by remote sensing, requires new techniques for their efficient and automated processing, including gap-filling of missing values. CUDA-based parallelization on GPU has become a popular way to dramatically increase computational efficiency of various approaches. Recently, we have proposed a computationally efficient and competitive, yet simple spatial prediction approach inspired from statistical physics models, called modified planar rotator (MPR) method. Its GPU implementation allowed additional impressive computational acceleration exceeding two orders of magnitude in comparison with CPU calculations. In the current study we propose a rather general approach to modelling spatial heterogeneity in GPU-implemented spatial prediction methods for two-dimensional gridded data by introducing spatial variability to model parameters. Predictions of unknown values are obtained from non-equilibrium conditional simulations, assuming ``local'' equilibrium conditions. We demonstrate that the proposed method leads to significant improvements in both prediction performance and computational efficiency.
|
stat-05.zip/2212.01317.pdf
|
2301.13749
|
Multi-Fidelity Covariance Estimation in the Log-Euclidean Geometry
|
[
"Aimee Maurais",
"Terrence Alsup",
"Benjamin Peherstorfer",
"Youssef Marzouk"
] |
31 Jan 2023 (<a href="https://arxiv.org/abs/2301.13749v1">v1</a>), last revised 26 May 2023 (this version, v2)
|
To appear at the International Conference on Machine Learning (ICML) 2023
|
Computation (stat.CO)
|
Computation (stat.CO); Machine Learning (cs.LG); Numerical Analysis (math.NA)
|
https://doi.org/10.48550/arXiv.2301.13749
|
We introduce a multi-fidelity estimator of covariance matrices that employs the log-Euclidean geometry of the symmetric positive-definite manifold. The estimator fuses samples from a hierarchy of data sources of differing fidelities and costs for variance reduction while guaranteeing definiteness, in contrast with previous approaches. The new estimator makes covariance estimation tractable in applications where simulation or data collection is expensive; to that end, we develop an optimal sample allocation scheme that minimizes the mean-squared error of the estimator given a fixed budget. Guaranteed definiteness is crucial to metric learning, data assimilation, and other downstream tasks. Evaluations of our approach using data from physical applications (heat conduction, fluid dynamics) demonstrate more accurate metric learning and speedups of more than one order of magnitude compared to benchmarks.
|
stat-05.zip/2301.13749.pdf
|
2308.10871
|
FunQuant: A R package to perform quantization in the context of rare events and time-consuming simulations
|
[
"Charlie Sire",
"Yann Richet",
"Rodolphe Le Riche",
"Didier Rullière",
"Jérémy Rohmer",
"Lucie Pheulpin"
] |
18 Aug 2023 (<a href="https://arxiv.org/abs/2308.10871v1">v1</a>), last revised 20 Sep 2024 (this version, v2)
|
7 pages, 4 figures. Submitted to Journal Of Open Source Software
|
Computation (stat.CO)
|
Computation (stat.CO); Machine Learning (cs.LG)
|
https://doi.org/10.48550/arXiv.2308.10871
|
Quantization summarizes continuous distributions by calculating a discrete approximation. Among the widely adopted methods for data quantization is Lloyd's algorithm, which partitions the space into Voronoï cells, that can be seen as clusters, and constructs a discrete distribution based on their centroids and probabilistic masses. Lloyd's algorithm estimates the optimal centroids in a minimal expected distance sense, but this approach poses significant challenges in scenarios where data evaluation is costly, and relates to rare events. Then, the single cluster associated to no event takes the majority of the probability mass. In this context, a metamodel is required and adapted sampling methods are necessary to increase the precision of the computations on the rare clusters.
|
stat-05.zip/2308.10871.pdf
|
2212.06748
|
A smoothed semiparametric likelihood for estimation of nonparametric finite mixture models with a copula-based dependence structure
|
[
"Michael Levine",
"Gildas Mazo"
] |
13 Dec 2022
|
Computation (stat.CO)
|
Computation (stat.CO)
|
https://doi.org/10.48550/arXiv.2212.06748
|
In this manuscript, we consider a finite multivariate nonparametric mixture model where the dependence between the marginal densities is modeled using the copula device. Pseudo EM stochastic algorithms were recently proposed to estimate all of the components of this model under a location-scale constraint on the marginals. Here, we introduce a deterministic algorithm that seeks to maximize a smoothed semiparametric likelihood. No location-scale assumption is made about the marginals. The algorithm is monotonic in one special case, and, in another, leads to ``approximate monotonicity'' -- whereby the difference between successive values of the objective function becomes non-negative up to an additive term that becomes negligible after a sufficiently large number of iterations. The behavior of this algorithm is illustrated on several simulated datasets. The results suggest that, under suitable conditions, the proposed algorithm may indeed be monotonic in general. A discussion of the results and some possible future research directions round out our presentation.
|
stat-05.zip/2212.06748.pdf
|
|
2302.11363
|
lqmix: an R package for longitudinal data analysis via linear quantile mixtures
|
[
"Marco Alfó",
"Maria Francesca Marino",
"Maria Giovanna Ranalli",
"Nicola Salvati"
] |
22 Feb 2023 (<a href="https://arxiv.org/abs/2302.11363v1">v1</a>), last revised 24 Mar 2025 (this version, v3)
|
25 pages, 2 figures
|
Computation (stat.CO)
|
Computation (stat.CO); Methodology (stat.ME)
|
https://doi.org/10.48550/arXiv.2302.11363
|
The analysis of longitudinal data gives the chance to observe how unit behaviors change over time, but it also poses a series of issues. These have been the focus of an extensive literature in the context of linear and generalized linear regression moving also, in the last ten years or so, to the context of linear quantile regression for continuous responses. In this paper, we present \textsf{lqmix}, a novel \textsf{R} package that assists in estimating a class of linear quantile regression models for longitudinal data, in the presence of time-constant and/or time-varying, unit-specific, random coefficients, with unspecified distribution. Model parameters are estimated in a maximum likelihood framework, via an extended EM algorithm, and parameters' standard errors are estimated via a block-bootstrap procedure. The analysis of a benchmark dataset is used to give details on the package functions.
|
stat-05.zip/2302.11363.pdf
|
2309.15735
|
Estimating MCMC convergence rates using common random number simulation
|
[
"Sabrina Sixta",
"Jeffrey S. Rosenthal",
"Austin Brown"
] |
27 Sep 2023 (<a href="https://arxiv.org/abs/2309.15735v1">v1</a>), last revised 29 May 2025 (this version, v3)
|
Computation (stat.CO)
|
Computation (stat.CO)
|
https://doi.org/10.48550/arXiv.2309.15735
|
This paper presents how to use common random number (CRN) simulation to evaluate Markov chain Monte Carlo (MCMC) convergence to stationarity. We provide an upper bound on the Wasserstein distance of a Markov chain to its stationary distribution after $N$ steps in terms of averages over CRN simulations. We apply our bound to Gibbs samplers on a variance component model, a model related to James-Stein estimators, and a Bayesian linear regression model. For the former two examples, we show that the CRN simulated bound converges to zero significantly more quickly compared to available drift and minorization bounds.
|
stat-05.zip/2309.15735.pdf
|
|
2309.17005
|
Bayesian Methodologies with pyhf
|
[
"Matthew Feickert",
"Lukas Heinrich",
"Malin Horstmann"
] |
29 Sep 2023 (<a href="https://arxiv.org/abs/2309.17005v1">v1</a>), last revised 11 Dec 2023 (this version, v2)
|
8 pages, 3 figures, 1 listing. Contribution to the Proceedings of the 26th International Conference on Computing In High Energy and Nuclear Physics (CHEP 2023)
|
Computation (stat.CO)
|
Computation (stat.CO); High Energy Physics - Experiment (hep-ex)
|
https://doi.org/10.48550/arXiv.2309.17005
|
bayesian_pyhf is a Python package that allows for the parallel Bayesian and frequentist evaluation of multi-channel binned statistical models. The Python library pyhf is used to build such models according to the HistFactory framework and already includes many frequentist inference methodologies. The pyhf-built models are then used as data-generating model for Bayesian inference and evaluated with the Python library PyMC. Based on Monte Carlo Chain Methods, PyMC allows for Bayesian modelling and together with the arviz library offers a wide range of Bayesian analysis tools.
|
stat-05.zip/2309.17005.pdf
|
2302.11801
|
Fast Computation of Branching Process Transition Probabilities via ADMM
|
[
"Achal Awasthi",
"Jason Xu"
] |
23 Feb 2023
|
32 pages, 4 figures, accepted by AISTATS 2023
|
Computation (stat.CO)
|
Computation (stat.CO); Applications (stat.AP)
|
https://doi.org/10.48550/arXiv.2302.11801
|
Branching processes are a class of continuous-time Markov chains (CTMCs) prevalent for modeling stochastic population dynamics in ecology, biology, epidemiology, and many other fields. The transient or finite-time behavior of these systems is fully characterized by their transition probabilities. However, computing them requires marginalizing over all paths between endpoint-conditioned values, which often poses a computational bottleneck. Leveraging recent results that connect generating function methods to a compressed sensing framework, we recast this task from the lens of sparse optimization. We propose a new solution method using variable splitting; in particular, we derive closed form updates in a highly efficient ADMM algorithm. Notably, no matrix products -- let alone inversions -- are required at any step. This reduces computational cost by orders of magnitude over existing methods, and the resulting algorithm is easily parallelizable and fairly insensitive to tuning parameters. A comparison to prior work is carried out in two applications to models of blood cell production and transposon evolution, showing that the proposed method is orders of magnitudes more scalable than existing work.
|
stat-05.zip/2302.11801.pdf
|
2301.10366
|
Construction of Uniform Designs over Continuous Domain in Computer Experiments
|
[
"Jianfa Lai",
"Kai-Tai Fang",
"Xiaoling Peng",
"Yuxuan Lin"
] |
25 Jan 2023
|
25 pages, 6 figures
|
Computation (stat.CO)
|
Computation (stat.CO)
|
https://doi.org/10.48550/arXiv.2301.10366
|
The construction of uniform designs (UDs) has received much attention in computer experiments over the past decades, but most of the previous works obtain uniform designs over a U-type by lattice domain. Due to increasing demands for continuous factors, UDs over the continuous domain is in lack. Moreover, the uniformity can be further improved over a continuous domain. In this paper, we use coordinate descent methods with an initialization derived by threshold accepting (TA) algorithm to construct UDs over a continuous domain with Centered $L_2$-discrepancy. The new UDs perform better than the recorded ones in several computer experiments by the Kriging modeling.
|
stat-05.zip/2301.10366.pdf
|
2309.01641
|
Expectation propagation for the smoothing distribution in dynamic probit
|
[
"Niccolò Anceschi",
"Augusto Fasano",
"Giovanni Rebaudo"
] |
4 Sep 2023
|
Computation (stat.CO)
|
Computation (stat.CO); Machine Learning (stat.ML)
|
https://doi.org/10.48550/arXiv.2309.01641
|
The smoothing distribution of dynamic probit models with Gaussian state dynamics was recently proved to belong to the unified skew-normal family. Although this is computationally tractable in small-to-moderate settings, it may become computationally impractical in higher dimensions. In this work, adapting a recent more general class of expectation propagation (EP) algorithms, we derive an efficient EP routine to perform inference for such a distribution. We show that the proposed approximation leads to accuracy gains over available approximate algorithms in a financial illustration.
|
stat-05.zip/2309.01641.pdf
|
|
2301.08297
|
Parametrization Cookbook: A set of Bijective Parametrizations for using Machine Learning methods in Statistical Inference
|
[
"Jean-Benoist Leger"
] |
19 Jan 2023
|
Computation (stat.CO)
|
Computation (stat.CO); Machine Learning (stat.ML)
|
https://doi.org/10.48550/arXiv.2301.08297
|
We present in this paper a way to transform a constrained statistical inference problem into an unconstrained one in order to be able to use modern computational methods, such as those based on automatic differentiation, GPU computing, stochastic gradients with mini-batch. <br>Unlike the parametrizations classically used in Machine Learning, the parametrizations introduced here are all bijective and are even diffeomorphisms, thus allowing to keep the important properties from a statistical inference point of view, first of all identifiability. <br>This cookbook presents a set of recipes to use to transform a constrained problem into a unconstrained one. <br>For an easy use of parametrizations, this paper is at the same time a cookbook, and a Python package allowing the use of parametrizations with numpy, but also JAX and PyTorch, as well as a high level and expressive interface allowing to easily describe a parametrization to transform a difficult problem of statistical inference into an easier problem addressable with modern optimization tools.
|
stat-05.zip/2301.08297.pdf
|
|
2301.01107
|
Computing the Performance of A New Adaptive Sampling Algorithm Based on The Gittins Index in Experiments with Exponential Rewards
|
[
"James K. He",
"Sofía S. Villar",
"Lida Mavrogonatou"
] |
3 Jan 2023
|
Accepted by Computing Conference, London 2023
|
Computation (stat.CO)
|
Computation (stat.CO); Machine Learning (cs.LG)
|
https://doi.org/10.48550/arXiv.2301.01107
|
Designing experiments often requires balancing between learning about the true treatment effects and earning from allocating more samples to the superior treatment. While optimal algorithms for the Multi-Armed Bandit Problem (MABP) provide allocation policies that optimally balance learning and earning, they tend to be computationally expensive. The Gittins Index (GI) is a solution to the MABP that can simultaneously attain optimality and computationally efficiency goals, and it has been recently used in experiments with Bernoulli and Gaussian rewards. For the first time, we present a modification of the GI rule that can be used in experiments with exponentially-distributed rewards. We report its performance in simulated 2- armed and 3-armed experiments. Compared to traditional non-adaptive designs, our novel GI modified design shows operating characteristics comparable in learning (e.g. statistical power) but substantially better in earning (e.g. direct benefits). This illustrates the potential that designs using a GI approach to allocate participants have to improve participant benefits, increase efficiencies, and reduce experimental costs in adaptive multi-armed experiments with exponential rewards.
|
stat-05.zip/2301.01107.pdf
|
2301.09516
|
Online Kernel Sliced Inverse Regression
|
[
"Wenquan Cui",
"Yue Zhao",
"Jianjun Xu",
"Haoyang Cheng"
] |
23 Jan 2023
|
Computation (stat.CO)
|
Computation (stat.CO); Statistics Theory (math.ST); Machine Learning (stat.ML)
|
https://doi.org/10.48550/arXiv.2301.09516
|
Online dimension reduction is a common method for high-dimensional streaming data processing. Online principal component analysis, online sliced inverse regression, online kernel principal component analysis and other methods have been studied in depth, but as far as we know, online supervised nonlinear dimension reduction methods have not been fully studied. In this article, an online kernel sliced inverse regression method is proposed. By introducing the approximate linear dependence condition and dictionary variable sets, we address the problem of increasing variable dimensions with the sample size in the online kernel sliced inverse regression method, and propose a reduced-order method for updating variables online. We then transform the problem into an online generalized eigen-decomposition problem, and use the stochastic optimization method to update the centered dimension reduction directions. Simulations and the real data analysis show that our method can achieve close performance to batch processing kernel sliced inverse regression.
|
stat-05.zip/2301.09516.pdf
|
|
2302.13398
|
Recursive Nearest Neighbor Co-Kriging Models for Big Multiple Fidelity Spatial Data Sets
|
[
"Si Cheng",
"Bledar A. Konomi",
"Georgos Karagiannis",
"Emily L. Kang"
] |
26 Feb 2023 (<a href="https://arxiv.org/abs/2302.13398v1">v1</a>), last revised 14 Nov 2023 (this version, v3)
|
Computation (stat.CO)
|
Computation (stat.CO)
|
https://doi.org/10.48550/arXiv.2302.13398
|
Big datasets are gathered daily from different remote sensing platforms. Recently, statistical co-kriging models, with the help of scalable techniques, have been able to combine such datasets by using spatially varying bias corrections. The associated Bayesian inference for these models is usually facilitated via Markov chain Monte Carlo (MCMC) methods which present (sometimes prohibitively) slow mixing and convergence because they require the simulation of high-dimensional random effect vectors from their posteriors given large datasets. To enable fast inference in big data spatial problems, we propose the recursive nearest neighbor co-kriging (RNNC) model. Based on this model, we develop two computationally efficient inferential procedures: a) the collapsed RNNC which reduces the posterior sampling space by integrating out the latent processes, and b) the conjugate RNNC, an MCMC free inference which significantly reduces the computational time without sacrificing prediction accuracy. An important highlight of conjugate RNNC is that it enables fast inference in massive multifidelity data sets by avoiding expensive integration algorithms. The efficient computational and good predictive performances of our proposed algorithms are demonstrated on benchmark examples and the analysis of the High-resolution Infrared Radiation Sounder data gathered from two NOAA polar orbiting satellites in which we managed to reduce the computational time from multiple hours to just a few minutes.
|
stat-05.zip/2302.13398.pdf
|
|
2302.12345
|
RESI: An R Package for Robust Effect Sizes
|
[
"Megan Jones",
"Kaidi Kang",
"Simon Vandekar"
] |
23 Feb 2023
|
25 pages, 2 figures. Submitted to Journal of Statistical Software
|
Computation (stat.CO)
|
Computation (stat.CO); Methodology (stat.ME)
|
https://doi.org/10.48550/arXiv.2302.12345
|
Effect size indices are useful parameters that quantify the strength of association and are unaffected by sample size. There are many available effect size parameters and estimators, but it is difficult to compare effect sizes across studies as most are defined for a specific type of population parameter. We recently introduced a new Robust Effect Size Index (RESI) and confidence interval, which is advantageous because it is not model-specific. Here we present the RESI R package, which makes it easy to report the RESI and its confidence interval for many different model classes, with a consistent interpretation across parameters and model types. The package produces coefficient, ANOVA tables, and overall Wald tests for model inputs, appending the RESI estimate and confidence interval to each. The package also includes functions for visualization and conversions to and from other effect size measures. For illustration, we analyze and interpret three different model types.
|
stat-05.zip/2302.12345.pdf
|
2301.09090
|
Parallel Approaches to Accelerate Bayesian Decision Trees
|
[
"Efthyvoulos Drousiotis",
"Paul G. Spirakis",
"Simon Maskell"
] |
22 Jan 2023
|
Computation (stat.CO)
|
Computation (stat.CO); Artificial Intelligence (cs.AI); Distributed, Parallel, and Cluster Computing (cs.DC); Machine Learning (cs.LG)
|
https://doi.org/10.48550/arXiv.2301.09090
|
Markov Chain Monte Carlo (MCMC) is a well-established family of algorithms primarily used in Bayesian statistics to sample from a target distribution when direct sampling is challenging. Existing work on Bayesian decision trees uses MCMC. Unfortunately, this can be slow, especially when considering large volumes of data. It is hard to parallelise the accept-reject component of the MCMC. None-the-less, we propose two methods for exploiting parallelism in the MCMC: in the first, we replace the MCMC with another numerical Bayesian approach, the Sequential Monte Carlo (SMC) sampler, which has the appealing property that it is an inherently parallel algorithm; in the second, we consider data partitioning. Both methods use multi-core processing with a HighPerformance Computing (HPC) resource. We test the two methods in various study settings to determine which method is the most beneficial for each test case. Experiments show that data partitioning has limited utility in the settings we consider and that the use of the SMC sampler can improve run-time (compared to the sequential implementation) by up to a factor of 343.
|
stat-05.zip/2301.09090.pdf
|
|
2302.13238
|
Statdepth: a package for analysis of functional and pointcloud data using statistical depth
|
[
"Julian Lehrer",
"Mircea Teodorescu",
"Yohei Rosen"
] |
26 Feb 2023
|
Computation (stat.CO)
|
Computation (stat.CO)
|
https://doi.org/10.48550/arXiv.2302.13238
|
Laboratory scientists are well equipped with statistical tools for univariate data, yet many phenomena of scientific interest are time-variant or otherwise multidimensional. Functional data analysis is one way of approaching such data: by representing these more complex data as single data points in a mathematical space of functions. The mathematical concept of functional depth provides a notion of centrality which allows for descriptive statistics and some comparative statistics on these data. Here, we present statdepth, a Python package for functional depth-based analyses which naturally extends familiar single-data-point L-statistics and related methods to time-variant data trajectories or multidimensional data.
|
stat-05.zip/2302.13238.pdf
|
|
2212.09931
|
A Generalized Variable Importance Metric and Estimator for Black Box Machine Learning Models
|
[
"Mohammad Kaviul Anam Khan",
"Olli Saarela",
"Rafal Kustra"
] |
20 Dec 2022 (<a href="https://arxiv.org/abs/2212.09931v1">v1</a>), last revised 23 Dec 2023 (this version, v3)
|
Computation (stat.CO)
|
Computation (stat.CO); Machine Learning (stat.ML)
|
https://doi.org/10.48550/arXiv.2212.09931
|
In this paper we define a population parameter, ``Generalized Variable Importance Metric (GVIM)'', to measure importance of predictors for black box machine learning methods, where the importance is not represented by model-based parameter. GVIM is defined for each input variable, using the true conditional expectation function, and it measures the variable's importance in affecting a continuous or a binary response. We extend previously published results to show that the defined GVIM can be represented as a function of the Conditional Average Treatment Effect (CATE) for any kind of a predictor, which gives it a causal interpretation and further justification as an alternative to classical measures of significance that are only available in simple parametric models. Extensive set of simulations using realistically complex relationships between covariates and outcomes and number of regression techniques of varying degree of complexity show the performance of our proposed estimator of the GVIM.
|
stat-05.zip/2212.09931.pdf
|
|
2212.12140
|
Design of Hamiltonian Monte Carlo for perfect simulation of general continuous distributions
|
[
"George M. Leigh",
"Amanda R. Northrop"
] |
23 Dec 2022
|
25 pages, 5 figures, ancillary material containing R scripts and data analysed
|
Computation (stat.CO)
|
Computation (stat.CO)
|
https://doi.org/10.48550/arXiv.2212.12140
|
Hamiltonian Monte Carlo (HMC) is an efficient method of simulating smooth distributions and has motivated the widely used No-U-turn Sampler (NUTS) and software Stan. We build on NUTS and the technique of "unbiased sampling" to design HMC algorithms that produce perfect simulation of general continuous distributions that are amenable to HMC. Our methods enable separation of Markov chain Monte Carlo convergence error from experimental error, and thereby provide much more powerful MCMC convergence diagnostics than current state-of-the-art summary statistics which confound these two errors. Objective comparison of different MCMC algorithms is provided by the number of derivative evaluations per perfect sample point. We demonstrate the methodology with applications to normal, $t$ and normal mixture distributions up to 100 dimensions, and a 12-dimensional Bayesian Lasso regression. HMC runs effectively with a goal of 20 to 30 points per trajectory. Numbers of derivative evaluations per perfect sample point range from 390 for a univariate normal distribution to 12,000 for a 100-dimensional mixture of two normal distributions with modes separated by six standard deviations, and 22,000 for a 100-dimensional $t$-distribution with four degrees of freedom.
|
stat-05.zip/2212.12140.pdf
|
2309.07882
|
Scalable Model-Based Gaussian Process Clustering
|
[
"Anirban Chakraborty",
"Abhisek Chakraborty"
] |
14 Sep 2023
|
Computation (stat.CO)
|
Computation (stat.CO); Applications (stat.AP); Machine Learning (stat.ML)
|
https://doi.org/10.48550/arXiv.2309.07882
|
Gaussian process is an indispensable tool in clustering functional data, owing to it's flexibility and inherent uncertainty quantification. However, when the functional data is observed over a large grid (say, of length $p$), Gaussian process clustering quickly renders itself infeasible, incurring $O(p^2)$ space complexity and $O(p^3)$ time complexity per iteration; and thus prohibiting it's natural adaptation to large environmental applications. To ensure scalability of Gaussian process clustering in such applications, we propose to embed the popular Vecchia approximation for Gaussian processes at the heart of the clustering task, provide crucial theoretical insights towards algorithmic design, and finally develop a computationally efficient expectation maximization (EM) algorithm. Empirical evidence of the utility of our proposal is provided via simulations and analysis of polar temperature anomaly (\href{<a href="https://www.ncei.noaa.gov/access/monitoring/climate-at-a-glance/global/time-series" rel="external noopener nofollow" class="link-external link-https">this https URL</a>}{<a href="http://noaa.gov" rel="external noopener nofollow" class="link-external link-http">this http URL</a>}) data-sets.
|
stat-05.zip/2309.07882.pdf
|
|
2309.05435
|
Parallel Selected Inversion for Space-Time Gaussian Markov Random Fields
|
[
"Abylay Zhumekenov",
"Elias T. Krainski",
"Håvard Rue"
] |
11 Sep 2023
|
17 pages, 7 figures
|
Computation (stat.CO)
|
Computation (stat.CO)
|
https://doi.org/10.48550/arXiv.2309.05435
|
Performing a Bayesian inference on large spatio-temporal models requires extracting inverse elements of large sparse precision matrices for marginal variances. Although direct matrix factorizations can be used for the inversion, such methods fail to scale well for distributed problems when run on large computing clusters. On the contrary, Krylov subspace methods for the selected inversion have been gaining traction. We propose a parallel hybrid approach based on domain decomposition, which extends the Rao-Blackwellized Monte Carlo estimator for distributed precision matrices. Our approach exploits the strength of Krylov subspace methods as global solvers and efficiency of direct factorizations as base case solvers to compute the marginal variances using a divide-and-conquer strategy. By introducing subdomain overlaps, one can achieve a greater accuracy at an increased computational effort with little to no additional communication. We demonstrate the speed improvements on both simulated models and a massive US daily temperature data.
|
stat-05.zip/2309.05435.pdf
|
2309.09367
|
ForLion: A New Algorithm for D-optimal Designs under General Parametric Statistical Models with Mixed Factors
|
[
"Yifei Huang",
"Keren Li",
"Abhyuday Mandal",
"Jie Yang"
] |
17 Sep 2023 (<a href="https://arxiv.org/abs/2309.09367v1">v1</a>), last revised 22 May 2024 (this version, v3)
|
36 pages, 7 tables, 5 figures
|
Computation (stat.CO)
|
Computation (stat.CO); Methodology (stat.ME)
|
https://doi.org/10.48550/arXiv.2309.09367
|
In this paper, we address the problem of designing an experimental plan with both discrete and continuous factors under fairly general parametric statistical models. We propose a new algorithm, named ForLion, to search for locally optimal approximate designs under the D-criterion. The algorithm performs an exhaustive search in a design space with mixed factors while keeping high efficiency and reducing the number of distinct experimental settings. Its optimality is guaranteed by the general equivalence theorem. We present the relevant theoretical results for multinomial logit models (MLM) and generalized linear models (GLM), and demonstrate the superiority of our algorithm over state-of-the-art design algorithms using real-life experiments under MLM and GLM. Our simulation studies show that the ForLion algorithm could reduce the number of experimental settings by 25% or improve the relative efficiency of the designs by 17.5% on average. Our algorithm can help the experimenters reduce the time cost, the usage of experimental devices, and thus the total cost of their experiments while preserving high efficiencies of the designs.
|
stat-05.zip/2309.09367.pdf
|
2303.01501
|
Stability and Machine Learning Applications of Persistent Homology Using the Delaunay-Rips Complex
|
[
"Amish Mishra",
"Francis C. Motta"
] |
2 Mar 2023
|
23 pages, 10 figures and tables
|
Computation (stat.CO)
|
Computation (stat.CO); Algebraic Topology (math.AT); Machine Learning (stat.ML)
|
https://doi.org/10.48550/arXiv.2303.01501
|
In this paper we define, implement, and investigate a simplicial complex construction for computing persistent homology of Euclidean point cloud data, which we call the Delaunay-Rips complex (DR). Assigning the Vietoris-Rips weights to simplices, DR experiences speed-up in the persistence calculations by only considering simplices that appear in the Delaunay triangulation of the point cloud. We document and compare a Python implementation of DR with other simplicial complex constructions for generating persistence diagrams. By imposing sufficient conditions on point cloud data, we are able to theoretically justify the stability of the persistence diagrams produced using DR. When the Delaunay triangulation of the point cloud changes under perturbations of the points, we prove that DR-produced persistence diagrams exhibit instability. Since we cannot guarantee that real-world data will satisfy our stability conditions, we demonstrate the practical robustness of DR for persistent homology in comparison with other simplicial complexes in machine learning applications. We find in our experiments that using DR for an ML-TDA pipeline performs comparatively well as using other simplicial complex constructions.
|
stat-05.zip/2303.01501.pdf
|
2309.01536
|
perms: Likelihood-free estimation of marginal likelihoods for binary response data in Python and R
|
[
"Dennis Christensen",
"Per August Jarval Moen"
] |
4 Sep 2023 (<a href="https://arxiv.org/abs/2309.01536v1">v1</a>), last revised 10 Dec 2023 (this version, v3)
|
Computation (stat.CO)
|
Computation (stat.CO); Methodology (stat.ME)
|
https://doi.org/10.48550/arXiv.2309.01536
|
In Bayesian statistics, the marginal likelihood (ML) is the key ingredient needed for model comparison and model averaging. Unfortunately, estimating MLs accurately is notoriously difficult, especially for models where posterior simulation is not possible. Recently, Christensen (2023) introduced the concept of permutation counting, which can accurately estimate MLs of models for exchangeable binary responses. Such data arise in a multitude of statistical problems, including binary classification, bioassay and sensitivity testing. Permutation counting is entirely likelihood-free and works for any model from which a random sample can be generated, including nonparametric models. Here we present perms, a package implementing permutation counting. As a result of extensive optimisation efforts, perms is computationally efficient and able to handle large data problems. It is available as both an R package and a Python library. A broad gallery of examples illustrating its usage is provided, which includes both standard parametric binary classification and novel applications of nonparametric models, such as changepoint analysis. We also cover the details of the implementation of perms and illustrate its computational speed via a simple simulation study.
|
stat-05.zip/2309.01536.pdf
|
|
2309.11455
|
ddtlcm: An R package for overcoming weak separation in Bayesian latent class analysis via tree-regularization
|
[
"Mengbing Li",
"Bolin Wu",
"Briana Stephenson",
"Zhenke Wu"
] |
20 Sep 2023
|
Computation (stat.CO)
|
Computation (stat.CO)
|
https://doi.org/10.48550/arXiv.2309.11455
|
Traditional applications of latent class models (LCMs) often focus on scenarios where a set of unobserved classes are well-defined and easily distinguishable. However, in numerous real-world applications, these classes are weakly separated and difficult to distinguish, creating significant numerical challenges. To address these issues, we have developed an R package ddtlcm that provides comprehensive analysis and visualization tools designed to enhance the robustness and interpretability of LCMs in the presence of weak class separation, particularly useful for small sample sizes. This package implements a tree-regularized Bayesian LCM that leverages statistical strength between latent classes to make better estimates using limited data. A Shiny app has also been developed to improve user interactivity. In this paper, we showcase a typical analysis pipeline with simulated data using ddtlcm. All software has been made publicly available on CRAN and GitHub.
|
stat-05.zip/2309.11455.pdf
|
|
2309.10589
|
Unbiased Parameter Estimation for Partially Observed Diffusions
|
[
"Elsiddig Awadelkarim",
"Ajay Jasra",
"Hamza Ruzayqat"
] |
19 Sep 2023
|
27 pages, 8 figures
|
Computation (stat.CO)
|
Computation (stat.CO); Numerical Analysis (math.NA); Methodology (stat.ME)
|
https://doi.org/10.48550/arXiv.2309.10589
|
In this article we consider the estimation of static parameters for partially observed diffusion process with discrete-time observations over a fixed time interval. In particular, we assume that one must time-discretize the partially observed diffusion process and work with the model with bias and consider maximizing the resulting log-likelihood. Using a novel double randomization scheme, based upon Markovian stochastic approximation we develop a new method to unbiasedly estimate the static parameters, that is, to obtain the maximum likelihood estimator with no time discretization bias. Under assumptions we prove that our estimator is unbiased and investigate the method in several numerical examples, showing that it can empirically out-perform existing unbiased methodology.
|
stat-05.zip/2309.10589.pdf
|
2301.02446
|
Optimal Scaling Results for Moreau-Yosida Metropolis-adjusted Langevin Algorithms
|
[
"Francesca R. Crucinio",
"Alain Durmus",
"Pablo Jiménez",
"Gareth O. Roberts"
] |
6 Jan 2023 (<a href="https://arxiv.org/abs/2301.02446v1">v1</a>), last revised 19 Jun 2024 (this version, v3)
|
Computation (stat.CO)
|
Computation (stat.CO); Probability (math.PR); Statistics Theory (math.ST)
|
https://doi.org/10.48550/arXiv.2301.02446
|
We consider a recently proposed class of MCMC methods which uses proximity maps instead of gradients to build proposal mechanisms which can be employed for both differentiable and non-differentiable targets. These methods have been shown to be stable for a wide class of targets, making them a valuable alternative to Metropolis-adjusted Langevin algorithms (MALA); and have found wide application in imaging contexts. The wider stability properties are obtained by building the Moreau-Yosida envelope for the target of interest, which depends on a parameter $\lambda$. In this work, we investigate the optimal scaling problem for this class of algorithms, which encompasses MALA, and provide practical guidelines for the implementation of these methods.
|
stat-05.zip/2301.02446.pdf
|
|
2309.14658
|
Improvements on Scalable Stochastic Bayesian Inference Methods for Multivariate Hawkes Process
|
[
"Alex Ziyu Jiang",
"Abel Rodríguez"
] |
26 Sep 2023 (<a href="https://arxiv.org/abs/2309.14658v1">v1</a>), last revised 21 Feb 2025 (this version, v3)
|
Computation (stat.CO)
|
Computation (stat.CO); Methodology (stat.ME)
|
https://doi.org/10.48550/arXiv.2309.14658
|
Multivariate Hawkes Processes (MHPs) are a class of point processes that can account for complex temporal dynamics among event sequences. In this work, we study the accuracy and computational efficiency of three classes of algorithms which, while widely used in the context of Bayesian inference, have rarely been applied in the context of MHPs: stochastic gradient expectation-maximization, stochastic gradient variational inference and stochastic gradient Langevin Monte Carlo. An important contribution of this paper is a novel approximation to the likelihood function that allows us to retain the computational advantages associated with conjugate settings while reducing approximation errors associated with the boundary effects. The comparisons are based on various simulated scenarios as well as an application to the study the risk dynamics in the Standard & Poor's 500 intraday index prices among its 11 sectors.
|
stat-05.zip/2309.14658.pdf
|
|
2302.10564
|
Computational issues in parameter estimation for hidden Markov models with Template Model Builder
|
[
"Timothée Bacri",
"Geir D. Berentsen",
"Jan Bulla",
"Bård Støve"
] |
21 Feb 2023 (<a href="https://arxiv.org/abs/2302.10564v1">v1</a>), last revised 15 May 2023 (this version, v2)
|
20 pages without appendix, 33 pages with appendix, 15 figures, 5 tables
|
Computation (stat.CO)
|
Computation (stat.CO)
|
https://doi.org/10.48550/arXiv.2302.10564
|
A popular way to estimate the parameters of a hidden Markov model (HMM) is direct numerical maximization (DNM) of the (log-)likelihood function. The advantages of employing the TMB (Kristensen et al., 2016) framework in R for this purpose were illustrated recently Bacri et al. (2022). In this paper, we present extensions of these results in two directions. First, we present a practical way to obtain uncertainty estimates in form of confidence intervals (CIs) for the so-called smoothing probabilities at moderate computational and programming effort via TMB. Our approach thus permits to avoid computer-intensive bootstrap methods. By means of several examples, we illustrate patterns present for the derived CIs. Secondly, we investigate the performance of popular optimizers available in R when estimating HMMs via DNM. Hereby, our focus lies on the potential benefits of employing TMB. Investigated criteria via a number of simulation studies are convergence speed, accuracy, and the impact of (poor) initial values. Our findings suggest that all optimizers considered benefit in terms of speed from using the gradient supplied by TMB. When supplying both gradient and Hessian from TMB, the number of iterations reduces, suggesting a more efficient convergence to the maximum of the log-likelihood. Last, we briefly point out potential advantages of a hybrid approach.
|
stat-05.zip/2302.10564.pdf
|
2309.14271
|
Generative Filtering for Recursive Bayesian Inference with Streaming Data
|
[
"Ian Taylor",
"Andee Kaplan",
"Brenda Betancourt"
] |
25 Sep 2023 (<a href="https://arxiv.org/abs/2309.14271v1">v1</a>), last revised 6 Dec 2024 (this version, v3)
|
Computation (stat.CO)
|
Computation (stat.CO)
|
https://doi.org/10.48550/arXiv.2309.14271
|
In the streaming data setting, where data arrive continuously or in frequent batches and there is no pre-determined amount of total data, Bayesian models can employ recursive updates, incorporating each new batch of data into the model parameters' posterior distribution. Filtering methods are currently used to perform these updates efficiently, however, they suffer from eventual degradation as the number of unique values within the filtered samples decreases. We propose Generative Filtering, a method for efficiently performing recursive Bayesian updates in the streaming setting. Generative Filtering retains the speed of a filtering method while using parallel updates to avoid degenerate distributions after repeated applications. We derive rates of convergence for Generative Filtering and conditions for the use of sufficient statistics instead of fully storing all past data. We investigate the alleviation of filtering degradation through simulation and Ecological species count data.
|
stat-05.zip/2309.14271.pdf
|
|
2310.00261
|
A physics and data co-driven surrogate modeling method for high-dimensional rare event simulation
|
[
"Jianhua Xian",
"Ziqi Wang"
] |
30 Sep 2023 (<a href="https://arxiv.org/abs/2310.00261v1">v1</a>), last revised 10 Apr 2024 (this version, v2)
|
Computation (stat.CO)
|
Computation (stat.CO); Applications (stat.AP)
|
https://doi.org/10.48550/arXiv.2310.00261
|
This paper presents a physics and data co-driven surrogate modeling method for efficient rare event simulation of civil and mechanical systems with high-dimensional input uncertainties. The method fuses interpretable low-fidelity physical models with data-driven error corrections. The hypothesis is that a well-designed and well-trained simplified physical model can preserve salient features of the original model, while data-fitting techniques can fill the remaining gaps between the surrogate and original model predictions. The coupled physics-data-driven surrogate model is adaptively trained using active learning, aiming to achieve a high correlation and small bias between the surrogate and original model responses in the critical parametric region of a rare event. A final importance sampling step is introduced to correct the surrogate model-based probability estimations. Static and dynamic problems with input uncertainties modeled by random field and stochastic process are studied to demonstrate the proposed method.
|
stat-05.zip/2310.00261.pdf
|
|
2309.12979
|
EgoCor: an R package to facilitate the use of exponential semi-variograms for modelling the local spatial correlation structure in social epidemiology
|
[
"Julia Dyck",
"Jan-Ole Koslik",
"Odile Sauzet"
] |
22 Sep 2023 (<a href="https://arxiv.org/abs/2309.12979v1">v1</a>), last revised 13 May 2024 (this version, v2)
|
Keywords: R package, semi-variogram, exponential models, small-area health inequalities
|
Computation (stat.CO)
|
Computation (stat.CO)
|
https://doi.org/10.48550/arXiv.2309.12979
|
As an alternative to using administrative areas for the evaluation of small-area health inequalities, Sauzet et al. suggested to take an ego-centred approach and model the spatial correlation structure of health outcomes at the individual level. Existing tools for the analysis of spatial data in R might appear too complex to non-specialists which could limit the use of the approach. We present the R package EgoCor which offers a user-friendly interface displaying in one function a range of graphics and tables of parameters to facilitate the decision making about which exponential parameters fit best either raw data or residuals. This function is based on the functions of the R package gstat. Moreover, we implemented a function providing the measure of uncertainty proposed by Dyck and Sauzet. With the R package EgoCor the modelling of spatial correlation structure of health outcomes or spatially structured predictors of health with a measure of uncertainty is made available to non-specialists.
|
stat-05.zip/2309.12979.pdf
|
2302.12546
|
Bayesian contiguity constrained clustering, spanning trees and dendrograms
|
[
"Etienne Côme"
] |
24 Feb 2023
|
Computation (stat.CO)
|
Computation (stat.CO); Methodology (stat.ME)
|
https://doi.org/10.48550/arXiv.2302.12546
|
Clustering is a well-known and studied problem, one of its variants, called contiguity-constrained clustering, accepts as a second input a graph used to encode prior information about cluster structure by means of contiguity constraints i.e. clusters must form connected subgraphs of this graph. This paper discusses the interest of such a setting and proposes a new way to formalise it in a Bayesian setting, using results on spanning trees to compute exactly a posteriori probabilities of candidate partitions. An algorithmic solution is then investigated to find a maximum a posteriori (MAP) partition and extract a Bayesian dendrogram from it. The interest of this last tool, which is reminiscent of the classical output of a simple hierarchical clustering algorithm, is analysed. Finally, the proposed approach is demonstrated with real applications. A reference implementation of this work is available in the R package gtclust that accompanies the paper (available at <a href="http://github.com/comeetie/gtclust" rel="external noopener nofollow" class="link-external link-http">this http URL</a>)
|
stat-05.zip/2302.12546.pdf
|
|
2310.01397
|
Posterior Uncertainty Estimation via a Monte Carlo Procedure Specialized for Data Assimilation
|
[
"Michael Stanley",
"Mikael Kuusela",
"Brendan Byrne",
"Junjie Liu"
] |
2 Oct 2023 (<a href="https://arxiv.org/abs/2310.01397v1">v1</a>), last revised 15 Nov 2023 (this version, v2)
|
Computation (stat.CO)
|
Computation (stat.CO)
|
https://doi.org/10.48550/arXiv.2310.01397
|
Through the Bayesian lens of data assimilation, uncertainty on model parameters is traditionally quantified through the posterior covariance matrix. However, in modern settings involving high-dimensional and computationally expensive forward models, posterior covariance knowledge must be relaxed to deterministic or stochastic approximations. In the carbon flux inversion literature, Chevallier et al. proposed a stochastic method capable of approximating posterior variances of linear functionals of the model parameters that is particularly well-suited for large-scale Earth-system data assimilation tasks. This note formalizes this algorithm and clarifies its properties. We provide a formal statement of the algorithm, demonstrate why it converges to the desired posterior variance quantity of interest, and provide additional uncertainty quantification allowing incorporation of the Monte Carlo sampling uncertainty into the method's Bayesian credible intervals. The methodology is demonstrated using toy simulations and a realistic carbon flux inversion observing system simulation experiment.
|
stat-05.zip/2310.01397.pdf
|
|
2302.13206
|
Semi-supervised Gaussian mixture modelling with a missing-data mechanism in R
|
[
"Ziyang Lyu",
"Daniel Ahfock",
"Ryan Thompson",
"Geoffrey J. McLachlan"
] |
26 Feb 2023 (<a href="https://arxiv.org/abs/2302.13206v1">v1</a>), last revised 17 Apr 2024 (this version, v3)
|
To appear in the Australian and New Zealand Journal of Statistics
|
Computation (stat.CO)
|
Computation (stat.CO)
|
https://doi.org/10.48550/arXiv.2302.13206
|
Semi-supervised learning is being extensively applied to estimate classifiers from training data in which not all the labels of the feature vectors are available. We present gmmsslm, an R package for estimating the Bayes' classifier from such partially classified data in the case where the feature vector has a multivariate Gaussian (normal) distribution in each of the predefined classes. Our package implements a recently proposed Gaussian mixture modelling framework that incorporates a missingness mechanism for the missing labels in which the probability of a missing label is represented via a logistic model with covariates that depend on the entropy of the feature vector. Under this framework, it has been shown that the accuracy of the Bayes' classifier formed from the Gaussian mixture model fitted to the partially classified training data can even have lower error rate than if it were estimated from the sample completely classified. This result was established in the particular case of two Gaussian classes with a common covariance matrix. Here, we focus on the effective implementation of an algorithm for multiple Gaussian classes with arbitrary covariance matrices. A strategy for initialising the algorithm is discussed and illustrated. The new package is demonstrated on some real data.
|
stat-05.zip/2302.13206.pdf
|
2310.01963
|
Quantifying the information lost in optimal covariance matrix cleaning
|
[
"Christian Bongiorno",
"Lamia Lamrani"
] |
3 Oct 2023 (<a href="https://arxiv.org/abs/2310.01963v1">v1</a>), last revised 8 Nov 2024 (this version, v2)
|
Computation (stat.CO)
|
Computation (stat.CO); Information Theory (cs.IT); Symbolic Computation (cs.SC)
|
https://doi.org/10.48550/arXiv.2310.01963
|
Obtaining an accurate estimate of the underlying covariance matrix from finite sample size data is challenging due to sample size noise. In recent years, sophisticated covariance-cleaning techniques based on random matrix theory have been proposed to address this issue. Most of these methods aim to achieve an optimal covariance matrix estimator by minimizing the Frobenius norm distance as a measure of the discrepancy between the true covariance matrix and the estimator. However, this practice offers limited interpretability in terms of information theory. To better understand this relationship, we focus on the Kullback-Leibler divergence to quantify the information lost by the estimator. Our analysis centers on rotationally invariant estimators, which are state-of-art in random matrix theory, and we derive an analytical expression for their Kullback-Leibler divergence. Due to the intricate nature of the calculations, we use genetic programming regressors paired with human intuition. Ultimately, using this approach, we formulate a conjecture validated through extensive simulations, showing that the Frobenius distance corresponds to a first-order expansion term of the Kullback-Leibler divergence, thus establishing a more defined link between the two measures.
|
stat-05.zip/2310.01963.pdf
|
|
2310.00125
|
Covariance Expressions for Multi-Fidelity Sampling with Multi-Output, Multi-Statistic Estimators: Application to Approximate Control Variates
|
[
"Thomas O. Dixon",
"James E. Warner",
"Geoffrey F. Bomarito",
"Alex A. Gorodetsky"
] |
29 Sep 2023 (<a href="https://arxiv.org/abs/2310.00125v1">v1</a>), last revised 28 Jun 2024 (this version, v2)
|
47 pages, 14 figures
|
Computation (stat.CO)
|
Computation (stat.CO); Methodology (stat.ME)
|
https://doi.org/10.48550/arXiv.2310.00125
|
We provide a collection of results on covariance expressions between Monte Carlo based multi-output mean, variance, and Sobol main effect variance estimators from an ensemble of models. These covariances can be used within multi-fidelity uncertainty quantification strategies that seek to reduce the estimator variance of high-fidelity Monte Carlo estimators with an ensemble of low-fidelity models. Such covariance expressions are required within approaches like the approximate control variate and multi-level best linear unbiased estimator. While the literature provides these expressions for some single-output cases such as mean and variance, our results are relevant to both multiple function outputs and multiple statistics across any sampling strategy. Following the description of these results, we use them within an approximate control variate scheme to show that leveraging multiple outputs can dramatically reduce estimator variance compared to single-output approaches. Synthetic examples are used to highlight the effects of optimal sample allocation and pilot sample estimation. A flight-trajectory simulation of entry, descent, and landing is used to demonstrate multi-output estimation in practical applications.
|
stat-05.zip/2310.00125.pdf
|
2303.00301
|
Auxiliary MCMC and particle Gibbs samplers for parallelisable inference in latent dynamical systems
|
[
"Adrien Corenflos",
"Simo Särkkä"
] |
1 Mar 2023 (<a href="https://arxiv.org/abs/2303.00301v1">v1</a>), last revised 3 Mar 2025 (this version, v3)
|
Computation (stat.CO)
|
Computation (stat.CO); Distributed, Parallel, and Cluster Computing (cs.DC); Machine Learning (stat.ML)
|
https://doi.org/10.48550/arXiv.2303.00301
|
Sampling from the full posterior distribution of high-dimensional non-linear, non-Gaussian latent dynamical models presents significant computational challenges. While Particle Gibbs (also known as conditional sequential Monte Carlo) is considered the gold standard for this task, it quickly degrades in performance as the latent space dimensionality increases. Conversely, globally Gaussian-approximated methods like extended Kalman filtering, though more robust, are seldom used for posterior sampling due to their inherent bias. We introduce novel auxiliary sampling approaches that address these limitations. By incorporating artificial observations of the system as auxiliary variables in our MCMC kernels, we develop both efficient exact Kalman-based samplers and enhanced Particle Gibbs algorithms that maintain performance in high-dimensional latent spaces. Some of our methods support parallelisation along the time dimension, achieving logarithmic scaling when implemented on GPUs. Empirical evaluations demonstrate superior statistical and computational performance compared to existing approaches for high-dimensional latent dynamical systems.
|
stat-05.zip/2303.00301.pdf
|
|
2303.03498
|
Properties of Marginal Sequential Monte Carlo Methods
|
[
"Francesca R. Crucinio",
"Adam M. Johansen"
] |
6 Mar 2023
|
Computation (stat.CO)
|
Computation (stat.CO); Probability (math.PR); Statistics Theory (math.ST); Methodology (stat.ME)
|
https://doi.org/10.48550/arXiv.2303.03498
|
We provide a framework which admits a number of ``marginal'' sequential Monte Carlo (SMC) algorithms as particular cases -- including the marginal particle filter [Klaas et al., 2005, in: Proceedings of Uncertainty in Artificial Intelligence, pp. 308--315], , the independent particle filter [Lin et al., 2005, Journal of the American Statistical Association 100, pp. 1412--1421] and linear-cost Approximate Bayesian Computation SMC [Sisson et al., 2007, Proceedings of the National Academy of Sciences (USA) 104, pp. 1760--1765.]. We provide conditions under which such algorithms obey laws of large numbers and central limit theorems and provide some further asymptotic characterizations. Finally, it is shown that the asymptotic variance of a class of estimators associated with certain marginal SMC algorithms is never greater than that of the estimators provided by a standard SMC algorithm using the same proposal distributions.
|
stat-05.zip/2303.03498.pdf
|
|
2308.13033
|
A Strength and Sparsity Preserving Algorithm for Generating Weighted, Directed Networks with Predetermined Assortativity
|
[
"Yelie Yuan",
"Jun Yan",
"Panpan Zhang"
] |
24 Aug 2023
|
Computation (stat.CO)
|
Computation (stat.CO)
|
https://doi.org/10.48550/arXiv.2308.13033
|
Degree-preserving rewiring is a widely used technique for generating unweighted networks with given assortativity, but for weighted networks, it is unclear how an analog would preserve the strengths and other critical network features such as sparsity level. This study introduces a novel approach for rewiring weighted networks to achieve desired directed assortativity. The method utilizes a mixed integer programming framework to establish a target network with predetermined assortativity coefficients, followed by an efficient rewiring algorithm termed "strength and sparsity preserving rewiring" (SSPR). SSPR retains the node strength distributions and network sparsity after rewiring. It is also possible to accommodate additional properties like edge weight distribution with extra computational cost. The optimization scheme can be used to determine feasible assortativity ranges for an initial network. The effectiveness of the proposed SSPR algorithm is demonstrated through its application to two classes of popular network models.
|
stat-05.zip/2308.13033.pdf
|
|
2310.00300
|
An Easy Rejection Sampling Baseline via Gradient Refined Proposals
|
[
"Edward Raff",
"Mark McLean",
"James Holt"
] |
30 Sep 2023
|
To appear in the 26th European Conference on Artificial Intelligence ECAI 2023
|
Computation (stat.CO)
|
Computation (stat.CO)
|
https://doi.org/10.48550/arXiv.2310.00300
|
Rejection sampling is a common tool for low dimensional problems ($d \leq 2$), often touted as an "easy" way to obtain valid samples from a distribution $f(\cdot)$ of interest. In practice it is non-trivial to apply, often requiring considerable mathematical effort to devise a good proposal distribution $g(\cdot)$ and select a supremum $C$. More advanced samplers require additional mathematical derivations, limitations on $f(\cdot)$, or even cross-validation, making them difficult to apply. We devise a new approximate baseline approach to rejection sampling that works with less information, requiring only a differentiable $f(\cdot)$ be specified, making it easier to use. We propose a new approach to rejection sampling by refining a parameterized proposal distribution with a loss derived from the acceptance threshold. In this manner we obtain comparable or better acceptance rates on current benchmarks by up to $7.3\times$, while requiring no extra assumptions or any derivations to use: only a differentiable $f(\cdot)$ is required. While approximate, the results are correct with high probability, and in all tests pass a distributional check. This makes our approach easy to use, reproduce, and efficacious.
|
stat-05.zip/2310.00300.pdf
|
2302.11449
|
From Optimization to Sampling Through Gradient Flows
|
[
"N. Garcia Trillos",
"B. Hosseini",
"D. Sanz-Alonso"
] |
22 Feb 2023
|
This article will appear in the Notices of the American Mathematical Society
|
Computation (stat.CO)
|
Computation (stat.CO)
|
https://doi.org/10.48550/arXiv.2302.11449
|
This article overviews how gradient flows, and discretizations thereof, are useful to design and analyze optimization and sampling algorithms. The interplay between optimization, sampling, and gradient flows is an active research area; our goal is to provide an accessible and lively introduction to some core ideas, emphasizing that gradient flows uncover the conceptual unity behind many optimization and sampling algorithms, and that they give a rich mathematical framework for their rigorous analysis.
|
stat-05.zip/2302.11449.pdf
|
2310.00561
|
CausalGPS: An R Package for Causal Inference With Continuous Exposures
|
[
"Naeem Khoshnevis",
"Xiao Wu",
"Danielle Braun"
] |
1 Oct 2023
|
22 pages, 8 figures
|
Computation (stat.CO)
|
Computation (stat.CO); Mathematical Software (cs.MS); Econometrics (econ.EM)
|
https://doi.org/10.48550/arXiv.2310.00561
|
Quantifying the causal effects of continuous exposures on outcomes of interest is critical for social, economic, health, and medical research. However, most existing software packages focus on binary exposures. We develop the CausalGPS R package that implements a collection of algorithms to provide algorithmic solutions for causal inference with continuous exposures. CausalGPS implements a causal inference workflow, with algorithms based on generalized propensity scores (GPS) as the core, extending propensity scores (the probability of a unit being exposed given pre-exposure covariates) from binary to continuous exposures. As the first step, the package implements efficient and flexible estimations of the GPS, allowing multiple user-specified modeling options. As the second step, the package provides two ways to adjust for confounding: weighting and matching, generating weighted and matched data sets, respectively. Lastly, the package provides built-in functions to fit flexible parametric, semi-parametric, or non-parametric regression models on the weighted or matched data to estimate the exposure-response function relating the outcome with the exposures. The computationally intensive tasks are implemented in C++, and efficient shared-memory parallelization is achieved by OpenMP API. This paper outlines the main components of the CausalGPS R package and demonstrates its application to assess the effect of long-term exposure to PM2.5 on educational attainment using zip code-level data from the contiguous United States from 2000-2016.
|
stat-05.zip/2310.00561.pdf
|
2306.11906
|
Statistical thinking in simulation design: a continuing conversation on the balancing intercept problem
|
[
"Boyi Guo",
"Linzi Li",
"Jacqueline E. Rudolph"
] |
20 Jun 2023
|
9 pages including title and abstract pages, references and supporting material, 1 figure
|
Computation (stat.CO)
|
Computation (stat.CO); Other Statistics (stat.OT)
|
https://doi.org/10.48550/arXiv.2306.11906
|
Epidemiologists have a growing interest in employing computational approaches to solve analytic problems, with simulation being arguably the most accessible among all approaches. While previous literature discussed the utility of simulation and demonstrated how to carry out them, few have focused on connecting underlying statistical concepts to these simulation approaches, creating gaps between theory and application. Based on the recent series of discussions on the balancing intercept, we explain the growing complexity when generalizing the balancing intercept to a wider class of simulations and revise the closed-form equation for the balancing intercept under assumptions. The discussion can broadly inform the future design of more complex simulations and emphasize the importance of applying statistical thinking in the new era of computational science.
|
stat-05.zip/2306.11906.pdf
|
2306.15841
|
On the surprising effectiveness of a simple matrix exponential derivative approximation, with application to global SARS-CoV-2
|
[
"Gustavo Didier",
"Nathan E. Glatt-Holtz",
"Andrew J. Holbrook",
"Andrew F. Magee",
"Marc A. Suchard"
] |
28 Jun 2023 (<a href="https://arxiv.org/abs/2306.15841v1">v1</a>), last revised 6 Dec 2023 (this version, v2)
|
To appear in the Proceedings of the National Academy of Sciences
|
Computation (stat.CO)
|
Computation (stat.CO); Probability (math.PR)
|
https://doi.org/10.48550/arXiv.2306.15841
|
The continuous-time Markov chain (CTMC) is the mathematical workhorse of evolutionary biology. Learning CTMC model parameters using modern, gradient-based methods requires the derivative of the matrix exponential evaluated at the CTMC's infinitesimal generator (rate) matrix. Motivated by the derivative's extreme computational complexity as a function of state space cardinality, recent work demonstrates the surprising effectiveness of a naive, first-order approximation for a host of problems in computational biology. In response to this empirical success, we obtain rigorous deterministic and probabilistic bounds for the error accrued by the naive approximation and establish a "blessing of dimensionality" result that is universal for a large class of rate matrices with random entries. Finally, we apply the first-order approximation within surrogate-trajectory Hamiltonian Monte Carlo for the analysis of the early spread of SARS-CoV-2 across 44 geographic regions that comprise a state space of unprecedented dimensionality for unstructured (flexible) CTMC models within evolutionary biology.
|
stat-05.zip/2306.15841.pdf
|
2306.15209
|
Dynamic Reconfiguration of Brain Functional Network in Stroke
|
[
"Kaichao Wu",
"Beth Jelfs",
"Katrina Neville",
"Wenzhen He",
"Qiang Fang"
] |
27 Jun 2023 (<a href="https://arxiv.org/abs/2306.15209v1">v1</a>), last revised 22 Mar 2024 (this version, v2)
|
Computation (stat.CO)
|
Computation (stat.CO); Applications (stat.AP); Methodology (stat.ME)
|
https://doi.org/10.1109/JBHI.2024.3371097
|
The brain continually reorganizes its functional network to adapt to post-stroke functional impairments. Previous studies using static modularity analysis have presented global-level behavior patterns of this network reorganization. However, it is far from understood how the brain reconfigures its functional network dynamically following a stroke. This study collected resting-state functional MRI data from 15 stroke patients, with mild (n = 6) and severe (n = 9) two subgroups based on their clinical symptoms. Additionally, 15 age-matched healthy subjects were considered as controls. By applying a multilayer network method, a dynamic modular structure was recognized based on a time-resolved function network. Then dynamic network measurements (recruitment, integration, and flexibility) were calculated to characterize the dynamic reconfiguration of post-stroke brain functional networks, hence, to reveal the neural functional rebuilding process. It was found from this investigation that severe patients tended to have reduced recruitment and increased between-network integration, while mild patients exhibited low network flexibility and less network integration. It is also noted that this severity-dependent alteration in network interaction was not able to be revealed by previous studies using static methods. Clinically, the obtained knowledge of the diverse patterns of dynamic adjustment in brain functional networks observed from the brain signal could help understand the underlying mechanism of the motor, speech, and cognitive functional impairments caused by stroke attacks. The proposed method not only could be used to evaluate patients' current brain status but also has the potential to provide insights into prognosis analysis and prediction.
|
stat-05.zip/2306.15209.pdf
|
|
2311.15539
|
A Novel Human-Based Meta-Heuristic Algorithm: Dragon Boat Optimization
|
[
"Xiang Li",
"Long Lan",
"Husam Lahza",
"Shaowu Yang",
"Shuihua Wang",
"Wenjing Yang",
"Hengzhu Liu",
"Yudong Zhang"
] |
27 Nov 2023
|
Computation (stat.CO)
|
Computation (stat.CO)
|
https://doi.org/10.48550/arXiv.2311.15539
|
(Aim) Dragon Boat Racing, a popular aquatic folklore team sport, is traditionally held during the Dragon Boat Festival. Inspired by this event, we propose a novel human-based meta-heuristic algorithm called dragon boat optimization (DBO) in this paper. (Method) It models the unique behaviors of each crew member on the dragon boat during the race by introducing social psychology mechanisms (social loafing, social incentive). Throughout this process, the focus is on the interaction and collaboration among the crew members, as well as their decision-making in different situations. During each iteration, DBO implements different state updating strategies. By modelling the crew's behavior and adjusting the state updating strategies, DBO is able to maintain high-performance efficiency. (Results) We have tested the DBO algorithm with 29 mathematical optimization problems and 2 structural design problems. (Conclusion) The experimental results demonstrate that DBO is competitive with state-of-the-art meta-heuristic algorithms as well as conventional methods.
|
stat-05.zip/2311.15539.pdf
|
|
2311.14492
|
Numerical Generalized Randomized HMC processes for restricted domains
|
[
"Tore Selland Kleppe",
"Roman Liesenfeld"
] |
24 Nov 2023 (<a href="https://arxiv.org/abs/2311.14492v1">v1</a>), last revised 31 May 2024 (this version, v2)
|
Computation (stat.CO)
|
Computation (stat.CO)
|
https://doi.org/10.48550/arXiv.2311.14492
|
We propose a generic approach for numerically efficient simulation from analytically intractable distributions with constrained support. Our approach relies upon Generalized Randomized Hamiltonian Monte Carlo (GRHMC) processes and combines these with a randomized transition kernel that appropriately adjusts the Hamiltonian flow at the boundary of the constrained domain, ensuring that it remains within the domain. The numerical implementation of this constrained GRHMC process exploits the sparsity of the randomized transition kernel and the specific structure of the constraints so that the proposed approach is numerically accurate, computationally fast and operational even in high-dimensional applications. We illustrate this approach with posterior distributions of several Bayesian models with challenging parameter domain constraints in applications to real-word data sets. Building on the capability of GRHMC processes to efficiently explore otherwise challenging and high-dimensional posteriors, the proposed method expands the set of Bayesian models that can be analyzed by using the standard Markov-Chain Monte-Carlo (MCMC) methodology, As such, it can advance the development and use of Bayesian models with useful constrained priors, which are difficult to handle with existing methods. The article is accompanied by an R-package (\url{<a href="https://github.com/torekleppe/pdmphmc" rel="external noopener nofollow" class="link-external link-https">this https URL</a>}), which allows for automatically implementing GRHMC processes for arbitrary target distributions and domain constraints.
|
stat-05.zip/2311.14492.pdf
|
|
2306.12184
|
A Note on the Relation Between Balenzela's Algorithm for Two-Filter Formula for Smoothing and Information Filter
|
[
"G. Kitagawa"
] |
21 Jun 2023
|
9 pages, 0 tables, 0 figures
|
Computation (stat.CO)
|
Computation (stat.CO)
|
https://doi.org/10.48550/arXiv.2306.12184
|
Recent paper by Balenzuela et al. presented an exact algorithm for computing the posterior distribution of current and future observations given the current state, $p(x_n|y_n,\ldots ,y_N)$, which is required when computing fixed-interval smoother of the state by a two-filter formula. In this note, it will be shown that their algorithm is equivalent to the backward filter obtained by applying an information filter to the reverse state-space model. Although their algorithm is proposed for complex Gaussian mixture distribution models, in this note, we consider the case of simple state-space models with respect to filter computation.
|
stat-05.zip/2306.12184.pdf
|
2311.17563
|
Efficient Computation of Sparse and Robust Maximum Association Estimators
|
[
"Pia Pfeiffer",
"Andreas Alfons",
"Peter Filzmoser"
] |
29 Nov 2023 (<a href="https://arxiv.org/abs/2311.17563v1">v1</a>), last revised 29 Jan 2025 (this version, v3)
|
Computation (stat.CO)
|
Computation (stat.CO); Machine Learning (stat.ML)
|
https://doi.org/10.48550/arXiv.2311.17563
|
Robust statistical estimators offer resilience against outliers but are often computationally challenging, particularly in high-dimensional sparse settings. Modern optimization techniques are utilized for robust sparse association estimators without imposing constraints on the covariance structure. The approach splits the problem into a robust estimation phase, followed by optimization of a decoupled, biconvex problem to derive the sparse canonical vectors. An augmented Lagrangian algorithm, combined with a modified adaptive gradient descent method, induces sparsity through simultaneous updates of both canonical vectors. Results demonstrate improved precision over existing methods, with high-dimensional empirical examples illustrating the effectiveness of this approach. The methodology can also be extended to other robust sparse estimators.
|
stat-05.zip/2311.17563.pdf
|
|
2311.17246
|
Detecting influential observations in single-index Fréchet regression
|
[
"Abdul-Nasah Soale"
] |
28 Nov 2023 (<a href="https://arxiv.org/abs/2311.17246v1">v1</a>), last revised 5 Jul 2024 (this version, v3)
|
Computation (stat.CO)
|
Computation (stat.CO); Applications (stat.AP); Methodology (stat.ME)
|
https://doi.org/10.48550/arXiv.2311.17246
|
Regression with random data objects is becoming increasingly common in modern data analysis. Unfortunately, this novel regression method is not immune to the trouble caused by unusual observations. A metric Cook's distance extending the original Cook's distances of Cook (1977) to regression between metric-valued response objects and Euclidean predictors is proposed. The performance of the metric Cook's distance is demonstrated in regression across four different response spaces in an extensive experimental study. Two real data applications involving the analyses of distributions of COVID-19 transmission in the State of Texas and the analyses of the structural brain connectivity networks are provided to illustrate the utility of the proposed method in practice.
|
stat-05.zip/2311.17246.pdf
|
|
2312.08823
|
Fast sampling from constrained spaces using the Metropolis-adjusted Mirror Langevin algorithm
|
[
"Vishwak Srinivasan",
"Andre Wibisono",
"Ashia Wilson"
] |
14 Dec 2023 (<a href="https://arxiv.org/abs/2312.08823v1">v1</a>), last revised 21 Jun 2024 (this version, v3)
|
49 pages, 6 figures, 2 tables. Shorter version without experiments accepted to COLT 2024
|
Computation (stat.CO)
|
Computation (stat.CO); Data Structures and Algorithms (cs.DS); Machine Learning (cs.LG); Statistics Theory (math.ST); Machine Learning (stat.ML)
|
https://doi.org/10.48550/arXiv.2312.08823
|
We propose a new method called the Metropolis-adjusted Mirror Langevin algorithm for approximate sampling from distributions whose support is a compact and convex set. This algorithm adds an accept-reject filter to the Markov chain induced by a single step of the Mirror Langevin algorithm (Zhang et al., 2020), which is a basic discretisation of the Mirror Langevin dynamics. Due to the inclusion of this filter, our method is unbiased relative to the target, while known discretisations of the Mirror Langevin dynamics including the Mirror Langevin algorithm have an asymptotic bias. For this algorithm, we also give upper bounds for the number of iterations taken to mix to a constrained distribution whose potential is relatively smooth, convex, and Lipschitz continuous with respect to a self-concordant mirror function. As a consequence of the reversibility of the Markov chain induced by the inclusion of the Metropolis-Hastings filter, we obtain an exponentially better dependence on the error tolerance for approximate constrained sampling. We also present numerical experiments that corroborate our theoretical findings.
|
stat-05.zip/2312.08823.pdf
|
2312.09380
|
Two-sample KS test with approxQuantile in Apache Spark
|
[
"Bradley Eck",
"Duygu Kabakci-Zorlu",
"Amadou Ba"
] |
14 Dec 2023
|
6 pages, 3 figures, python code in appendix. To appear at IEEE Big Data 2023
|
Computation (stat.CO)
|
Computation (stat.CO)
|
https://doi.org/10.48550/arXiv.2312.09380
|
The classical two-sample test of Kolmogorov-Smirnov (KS) is widely used to test whether empirical samples come from the same distribution. Even though most statistical packages provide an implementation, carrying out the test in big data settings can be challenging because it requires a full sort of the data. The popular Apache Spark system for big data processing provides a 1-sample KS test, but not the 2-sample version. Moreover, recent Spark versions provide the approxQuantile method for querying $\epsilon$-approximate quantiles. We build on approxQuantile to propose a variation of the classical Kolmogorov-Smirnov two-sample test that constructs approximate cumulative distribution functions (CDF) from $\epsilon$-approximate quantiles. We derive error bounds of the approximate CDF and show how to use this information to carry out KS tests. Psuedocode for the approach requires 15 executable lines. A Python implementation appears in the appendix.
|
stat-05.zip/2312.09380.pdf
|
2312.03538
|
Bayesian variable selection in sample selection models using spike-and-slab priors
|
[
"Adam Iqbal",
"Emmanuel O. Ogundimu",
"F. Javier Rubio"
] |
6 Dec 2023 (<a href="https://arxiv.org/abs/2312.03538v1">v1</a>), last revised 21 Jan 2025 (this version, v3)
|
Computation (stat.CO)
|
Computation (stat.CO); Methodology (stat.ME)
|
https://doi.org/10.48550/arXiv.2312.03538
|
Sample selection models represent a common methodology for correcting bias induced by data missing not at random. These models are not empirically identifiable without exclusion restrictions. In other words, some variables predictive of missingness do not affect the outcome model of interest. The drive to establish this requirement often leads to the inclusion of irrelevant variables in the model. A recent proposal uses adaptive LASSO to circumvent this problem, but its performance depends on the so-called covariance assumption, which can be violated in small to moderate samples. Additionally, there are no tools yet for post-selection inference for this model. To address these challenges, we propose two families of spike-and-slab priors to conduct Bayesian variable selection in sample selection models. These prior structures allow for constructing a Gibbs sampler with tractable conditionals, which is scalable to the dimensions of practical interest. We illustrate the performance of the proposed methodology through a simulation study and present a comparison against adaptive LASSO and stepwise selection. We also provide two applications using publicly available real data. An implementation and code to reproduce the results in this paper can be found at <a href="https://github.com/adam-iqbal/selection-spike-slab" rel="external noopener nofollow" class="link-external link-https">this https URL</a>.
|
stat-05.zip/2312.03538.pdf
|
|
2312.06484
|
Augmented Island Resampling Particle Filters for Particle Markov Chain Monte Carlo
|
[
"Kari Heine"
] |
11 Dec 2023
|
Computation (stat.CO)
|
Computation (stat.CO)
|
https://doi.org/10.48550/arXiv.2312.06484
|
In modern days, the ability to carry out computations in parallel is key to efficient implementations of computationally intensive algorithms. This paper investigates the applicability of the previously proposed Augmented Island Resampling Particle Filter (AIRPF) -- an algorithm designed for parallel implementations -- to particle Markov Chain Monte Carlo (PMCMC). We show that AIRPF produces a non-negative unbiased estimator of the marginal likelihood and hence is suitable for PMCMC. We also prove stability properties, similar to those of the $\alpha$SMC algorithm, for AIRPF. This implies that the error of AIRPF can be bounded uniformly in time by controlling the effective number of filters, which in turn can be done by adaptively constraining the interactions between filters. We demonstrate the superiority of AIRPF over independent Bootstrap Particle Filters, not only numerically, but also theoretically. To this end, we extend the previously proposed collision analysis approach to derive an explicit expression for the variance of the marginal likelihood estimate. This expression admits exact evaluation of the variance in some simple scenarios as we shall also demonstrate.
|
stat-05.zip/2312.06484.pdf
|
|
2312.04898
|
Quantifying the effectiveness of linear preconditioning in Markov chain Monte Carlo
|
[
"Max Hird",
"Samuel Livingstone"
] |
8 Dec 2023 (<a href="https://arxiv.org/abs/2312.04898v1">v1</a>), last revised 4 Dec 2024 (this version, v2)
|
Computation (stat.CO)
|
Computation (stat.CO); Statistics Theory (math.ST); Other Statistics (stat.OT)
|
https://doi.org/10.48550/arXiv.2312.04898
|
We study linear preconditioning in Markov chain Monte Carlo. We consider the class of well-conditioned distributions, for which several mixing time bounds depend on the condition number $\kappa$. First we show that well-conditioned distributions exist for which $\kappa$ can be arbitrarily large and yet no linear preconditioner can reduce it. We then impose two sets of extra assumptions under which a linear preconditioner can significantly reduce $\kappa$. For the random walk Metropolis we further provide upper and lower bounds on the spectral gap with tight $1/\kappa$ dependence. This allows us to give conditions under which linear preconditioning can provably increase the gap. We then study popular preconditioners such as the covariance, its diagonal approximation, the hessian at the mode, and the QR decomposition. We show conditions under which each of these reduce $\kappa$ to near its minimum. We also show that the diagonal approach can in fact \textit{increase} the condition number. This is of interest as diagonal preconditioning is the default choice in well-known software packages. We conclude with a numerical study comparing preconditioners in different models, and showing how proper preconditioning can greatly reduce compute time in Hamiltonian Monte Carlo.
|
stat-05.zip/2312.04898.pdf
|
|
2306.06405
|
Effects of 3D Position Fluctuations on Air-to-Ground mmWave UAV Communications
|
[
"Cunyan Ma",
"Xiaoya Li",
"Chen He",
"Jinye Peng",
"Z. Jane Wang"
] |
10 Jun 2023
|
Computation (stat.CO)
|
Computation (stat.CO)
|
https://doi.org/10.48550/arXiv.2306.06405
|
Millimeter wave (mmWave)-based unmanned aerial vehicle (UAV) communication is a promising candidate for future communications due to its flexibility and sufficient bandwidth. However, random fluctuations in the position of hovering UAVs will lead to random variations in the blockage and signal-to-noise ratio (SNR) of the UAV-user link, thus affecting the quality of service (QoS) of the system. To assess the impact of UAV position fluctuations on the QoS of air-to-ground mmWave UAV communications, this paper develops a tractable analytical model that jointly captures the features of three-dimensional (3D) position fluctuations of hovering UAVs and blockages of mmWave (including static, dynamic, and self-blockages). With this model, we derive the closed-form expressions for reliable service probability respective to blockage probability of UAV-user links,and coverage probability respective to SNR, respectively. The results indicate that the greater the position fluctuations of UAVs, the lower the reliable service probability and coverage probability. The degradation of these two evaluation metrics confirms that the performance of air-to-ground mmWave UAV systems largely depends on the UAV position fluctuations, and the stronger the fluctuation, the worse the QoS. Finally, Monte Carlo simulations demonstrate the above results and show UAVs' optimal location to maximize the reliable service and coverage probability, respectively.
|
stat-05.zip/2306.06405.pdf
|
|
2306.04798
|
A Trigamma-free Approach for Computing Information Matrices Related to Trigamma Function
|
[
"Zhou Yu",
"Niloufar Dousti Mousavi",
"Jie Yang"
] |
7 Jun 2023 (<a href="https://arxiv.org/abs/2306.04798v1">v1</a>), last revised 19 Jan 2024 (this version, v2)
|
2 figures, 9 tables
|
Computation (stat.CO)
|
Computation (stat.CO)
|
https://doi.org/10.48550/arXiv.2306.04798
|
Negative binomial related distributions have been widely used in practice. The calculation of the corresponding Fisher information matrices involves the expectation of trigamma function values which can only be calculated numerically and approximately. In this paper, we propose a trigamma-free approach to approximate the expectations involving the trigamma function, along with theoretical upper bounds for approximation errors. We show by numerical studies that our approach is highly efficient and much more accurate than previous methods. We also apply our approach to compute the Fisher information matrices of zero-inflated negative binomial (ZINB) and beta negative binomial (ZIBNB) probabilistic models, as well as ZIBNB regression models.
|
stat-05.zip/2306.04798.pdf
|
2312.08728
|
Mini-batch Gradient Descent with Buffer
|
[
"Haobo Qi",
"Du Huang",
"Yingqiu Zhu",
"Danyang Huang",
"Hansheng Wang"
] |
14 Dec 2023
|
Computation (stat.CO)
|
Computation (stat.CO)
|
https://doi.org/10.48550/arXiv.2312.08728
|
In this paper, we studied a buffered mini-batch gradient descent (BMGD) algorithm for training complex model on massive datasets. The algorithm studied here is designed for fast training on a GPU-CPU system, which contains two steps: the buffering step and the computation step. In the buffering step, a large batch of data (i.e., a buffer) are loaded from the hard drive to the graphical memory of GPU. In the computation step, a standard mini-batch gradient descent(MGD) algorithm is applied to the buffered data. Compared to traditional MGD algorithm, the proposed BMGD algorithm can be more efficient for two <a href="http://reasons.First" rel="external noopener nofollow" class="link-external link-http">this http URL</a>, the BMGD algorithm uses the buffered data for multiple rounds of gradient update, which reduces the expensive communication cost from the hard drive to GPU memory. Second, the buffering step can be executed in parallel so that the GPU does not have to stay idle when loading new data. We first investigate the theoretical properties of BMGD algorithms under a linear regression <a href="http://setting.The" rel="external noopener nofollow" class="link-external link-http">this http URL</a> analysis is then extended to the Polyak-Lojasiewicz loss function class. The theoretical claims about the BMGD algorithm are numerically verified by simulation studies. The practical usefulness of the proposed method is demonstrated by three image-related real data analysis.
|
stat-05.zip/2312.08728.pdf
|
|
2306.04093
|
Subnetwork Estimation for Spatial Autoregressive Models in Large-scale Networks
|
[
"Xuetong Li",
"Feifei Wang",
"Wei Lan",
"Hansheng Wang"
] |
7 Jun 2023 (<a href="https://arxiv.org/abs/2306.04093v1">v1</a>), last revised 9 Jun 2023 (this version, v2)
|
Computation (stat.CO)
|
Computation (stat.CO)
|
https://doi.org/10.48550/arXiv.2306.04093
|
Large-scale networks are commonly encountered in practice (e.g., Facebook and Twitter) by researchers. In order to study the network interaction between different nodes of large-scale networks, the spatial autoregressive (SAR) model has been popularly employed. Despite its popularity, the estimation of a SAR model on large-scale networks remains very challenging. On the one hand, due to policy limitations or high collection costs, it is often impossible for independent researchers to observe or collect all network information. On the other hand, even if the entire network is accessible, estimating the SAR model using the quasi-maximum likelihood estimator (QMLE) could be computationally infeasible due to its high computational cost. To address these challenges, we propose here a subnetwork estimation method based on QMLE for the SAR model. By using appropriate sampling methods, a subnetwork, consisting of a much-reduced number of nodes, can be constructed. Subsequently, the standard QMLE can be computed by treating the sampled subnetwork as if it were the entire network. This leads to a significant reduction in information collection and model computation costs, which increases the practical feasibility of the effort. Theoretically, we show that the subnetwork-based QMLE is consistent and asymptotically normal under appropriate regularity conditions. Extensive simulation studies, based on both simulated and real network structures, are presented.
|
stat-05.zip/2306.04093.pdf
|
|
2306.06992
|
Extending JumpProcess.jl for fast point process simulation with time-varying intensities
|
[
"Guilherme Augusto Zagatti",
"Samuel A. Isaacson",
"Christopher Rackauckas",
"Vasily Ilin",
"See-Kiong Ng",
"Stéphane Bressan"
] |
12 Jun 2023 (<a href="https://arxiv.org/abs/2306.06992v1">v1</a>), last revised 24 Jul 2023 (this version, v3)
|
Computation (stat.CO)
|
Computation (stat.CO); Mathematical Software (cs.MS)
|
https://doi.org/10.48550/arXiv.2306.06992
|
Point processes model the occurrence of a countable number of random points over some support. They can model diverse phenomena, such as chemical reactions, stock market transactions and social interactions. We show that <a href="http://JumpProcesses.jl" rel="external noopener nofollow" class="link-external link-http">this http URL</a> is a fast, general-purpose library for simulating point processes. <a href="http://JumpProcesses.jl" rel="external noopener nofollow" class="link-external link-http">this http URL</a> was first developed for simulating jump processes via stochastic simulation algorithms (SSAs) (including Doob's method, Gillespie's methods, and Kinetic Monte Carlo methods). Historically, jump processes have been developed in the context of dynamical systems to describe dynamics with discrete jumps. In contrast, the development of point processes has been more focused on describing the occurrence of random events. In this paper, we bridge the gap between the treatment of point and jump process simulation. The algorithms previously included in <a href="http://JumpProcesses.jl" rel="external noopener nofollow" class="link-external link-http">this http URL</a> can be mapped to three general methods developed in statistics for simulating evolutionary point processes. Our comparative exercise revealed that the library initially lacked an efficient algorithm for simulating processes with variable intensity rates. We, therefore, extended <a href="http://JumpProcesses.jl" rel="external noopener nofollow" class="link-external link-http">this http URL</a> with a new simulation algorithm, Coevolve, that enables the rapid simulation of processes with locally-bounded variable intensity rates. It is now possible to efficiently simulate any point process on the real line with a non-negative, left-continuous, history-adapted and locally bounded intensity rate coupled or not with differential equations. This extension significantly improves the computational performance of <a href="http://JumpProcesses.jl" rel="external noopener nofollow" class="link-external link-http">this http URL</a> when simulating such processes, enabling it to become one of the few readily available, fast, general-purpose libraries for simulating evolutionary point processes.
|
stat-05.zip/2306.06992.pdf
|
|
2306.00389
|
R-VGAL: A Sequential Variational Bayes Algorithm for Generalised Linear Mixed Models
|
[
"Bao Anh Vu",
"David Gunawan",
"Andrew Zammit-Mangion"
] |
1 Jun 2023 (<a href="https://arxiv.org/abs/2306.00389v1">v1</a>), last revised 18 Apr 2024 (this version, v2)
|
Computation (stat.CO)
|
Computation (stat.CO); Methodology (stat.ME)
|
https://doi.org/10.48550/arXiv.2306.00389
|
Models with random effects, such as generalised linear mixed models (GLMMs), are often used for analysing clustered data. Parameter inference with these models is difficult because of the presence of cluster-specific random effects, which must be integrated out when evaluating the likelihood function. Here, we propose a sequential variational Bayes algorithm, called Recursive Variational Gaussian Approximation for Latent variable models (R-VGAL), for estimating parameters in GLMMs. The R-VGAL algorithm operates on the data sequentially, requires only a single pass through the data, and can provide parameter updates as new data are collected without the need of re-processing the previous data. At each update, the R-VGAL algorithm requires the gradient and Hessian of a "partial" log-likelihood function evaluated at the new observation, which are generally not available in closed form for GLMMs. To circumvent this issue, we propose using an importance-sampling-based approach for estimating the gradient and Hessian via Fisher's and Louis' identities. We find that R-VGAL can be unstable when traversing the first few data points, but that this issue can be mitigated by using a variant of variational tempering in the initial steps of the algorithm. Through illustrations on both simulated and real datasets, we show that R-VGAL provides good approximations to the exact posterior distributions, that it can be made robust through tempering, and that it is computationally efficient.
|
stat-05.zip/2306.00389.pdf
|
|
2306.01575
|
BayesMortalityPlus: A package in R for Bayesian graduation of mortality modelling
|
[
"Lucas M. F. Silva",
"Luiz F. V. Figueiredo",
"Viviana G. R. Lobo",
"Thaís C. O. Fonseca",
"Mariane B. Alves"
] |
2 Jun 2023
|
Computation (stat.CO)
|
Computation (stat.CO); Applications (stat.AP)
|
https://doi.org/10.48550/arXiv.2306.01575
|
The BayesMortalityPlus package provides a framework for modelling and predicting mortality data. The package includes tools for the construction of life tables based on Heligman-Pollard laws, and also on dynamic linear smoothers. Flexibility is available in terms of modelling so that the response variable may be modeled as Poisson, Binomial or Gaussian. If temporal data is available, the package provides a Bayesian implementation for the well-known Lee-Carter model that allows for estimation, projection of mortality over time, and assessment of uncertainty of any linear or nonlinear function of parameters such as life expectancy. Illustrations are considered to show the capability of the proposed package to model mortality data.
|
stat-05.zip/2306.01575.pdf
|
|
2309.01619
|
Efficient expectation propagation for posterior approximation in high-dimensional probit models
|
[
"Augusto Fasano",
"Niccolò Anceschi",
"Beatrice Franzolini",
"Giovanni Rebaudo"
] |
4 Sep 2023
|
Computation (stat.CO)
|
Computation (stat.CO); Methodology (stat.ME); Machine Learning (stat.ML)
|
https://doi.org/10.48550/arXiv.2309.01619
|
Bayesian binary regression is a prosperous area of research due to the computational challenges encountered by currently available methods either for high-dimensional settings or large datasets, or both. In the present work, we focus on the expectation propagation (EP) approximation of the posterior distribution in Bayesian probit regression under a multivariate Gaussian prior distribution. Adapting more general derivations in Anceschi et al. (2023), we show how to leverage results on the extended multivariate skew-normal distribution to derive an efficient implementation of the EP routine having a per-iteration cost that scales linearly in the number of covariates. This makes EP computationally feasible also in challenging high-dimensional settings, as shown in a detailed simulation study.
|
stat-05.zip/2309.01619.pdf
|
|
2308.14952
|
Stochastic Variational Inference for GARCH Models
|
[
"Hanwen Xuan",
"Luca Maestrini",
"Feng Chen",
"Clara Grazian"
] |
29 Aug 2023
|
23 pages, 10 figures
|
Computation (stat.CO)
|
Computation (stat.CO); Econometrics (econ.EM); Applications (stat.AP)
|
https://doi.org/10.48550/arXiv.2308.14952
|
Stochastic variational inference algorithms are derived for fitting various heteroskedastic time series models. We examine Gaussian, t, and skew-t response GARCH models and fit these using Gaussian variational approximating densities. We implement efficient stochastic gradient ascent procedures based on the use of control variates or the reparameterization trick and demonstrate that the proposed implementations provide a fast and accurate alternative to Markov chain Monte Carlo sampling. Additionally, we present sequential updating versions of our variational algorithms, which are suitable for efficient portfolio construction and dynamic asset allocation.
|
stat-05.zip/2308.14952.pdf
|
2309.02998
|
Multilevel Particle Filters for a Class of Partially Observed Piecewise Deterministic Markov Processes
|
[
"Ajay Jasra",
"Kengo Kamatani",
"Mohamed Maama"
] |
6 Sep 2023
|
Computation (stat.CO)
|
Computation (stat.CO)
|
https://doi.org/10.48550/arXiv.2309.02998
|
In this paper we consider the filtering of a class of partially observed piecewise deterministic Markov processes (PDMPs). In particular, we assume that an ordinary differential equation (ODE) drives the deterministic element and can only be solved numerically via a time discretization. We develop, based upon the approach in [20], a new particle and multilevel particle filter (MLPF) in order to approximate the filter associated to the discretized ODE. We provide a bound on the mean square error associated to the MLPF which provides guidance on setting the simulation parameter of that algorithm and implies that significant computational gains can be obtained versus using a particle filter. Our theoretical claims are confirmed in several numerical examples.
|
stat-05.zip/2309.02998.pdf
|
|
2309.01630
|
Efficient computation of predictive probabilities in probit models via expectation propagation
|
[
"Augusto Fasano",
"Niccolò Anceschi",
"Beatrice Franzolini",
"Giovanni Rebaudo"
] |
4 Sep 2023
|
Computation (stat.CO)
|
Computation (stat.CO); Methodology (stat.ME); Machine Learning (stat.ML)
|
https://doi.org/10.48550/arXiv.2309.01630
|
Binary regression models represent a popular model-based approach for binary classification. In the Bayesian framework, computational challenges in the form of the posterior distribution motivate still-ongoing fruitful research. Here, we focus on the computation of predictive probabilities in Bayesian probit models via expectation propagation (EP). Leveraging more general results in recent literature, we show that such predictive probabilities admit a closed-form expression. Improvements over state-of-the-art approaches are shown in a simulation study.
|
stat-05.zip/2309.01630.pdf
|
|
2401.00945
|
A review of Monte Carlo-based versions of the EM algorithm
|
[
"William Ruth"
] |
1 Jan 2024
|
Computation (stat.CO)
|
Computation (stat.CO); Methodology (stat.ME)
|
https://doi.org/10.48550/arXiv.2401.00945
|
The EM algorithm is a powerful tool for maximum likelihood estimation with missing data. In practice, the calculations required for the EM algorithm are often intractable. We review numerous methods to circumvent this intractability, all of which are based on Monte Carlo simulation. We focus our attention on the Monte Carlo EM (MCEM) algorithm and its various implementations. We also discuss some related methods like stochastic approximation and Monte Carlo maximum likelihood. Generating the Monte Carlo samples necessary for these methods is, in general, a hard problem. As such, we review several simulation strategies which can be used to address this challenge. <br>Given the wide range of methods available for approximating the EM, it can be challenging to select which one to use. We review numerous comparisons between these methods from a wide range of sources, and offer guidance on synthesizing the findings. Finally, we give some directions for future research to fill important gaps in the existing literature on the MCEM algorithm and related methods.
|
stat-05.zip/2401.00945.pdf
|
|
2305.17562
|
Mixed-integer linear programming for computing optimal experimental designs
|
[
"Radoslav Harman",
"Samuel Rosa"
] |
27 May 2023 (<a href="https://arxiv.org/abs/2305.17562v1">v1</a>), last revised 16 Jun 2024 (this version, v2)
|
Accepted manuscript
|
Computation (stat.CO)
|
Computation (stat.CO)
|
https://doi.org/10.48550/arXiv.2305.17562
|
The problem of computing an exact experimental design that is optimal for the least-squares estimation of the parameters of a regression model is considered. We show that this problem can be solved via mixed-integer linear programming (MILP) for a wide class of optimality criteria, including the criteria of A-, I-, G- and MV-optimality. This approach improves upon the current state-of-the-art mathematical programming formulation, which uses mixed-integer second-order cone programming. The key idea underlying the MILP formulation is McCormick relaxation, which critically depends on finite interval bounds for the elements of the covariance matrix of the least-squares estimator corresponding to an optimal exact design. We provide both analytic and algorithmic methods for constructing these bounds. We also demonstrate the unique advantages of the MILP approach, such as the possibility of incorporating multiple design constraints into the optimization problem, including constraints on the variances and covariances of the least-squares estimator.
|
stat-05.zip/2305.17562.pdf
|
2306.00469
|
HiQR: An efficient algorithm for high-dimensional quadratic regression with penalties
|
[
"Cheng Wang",
"Haozhe Chen",
"Binyan Jiang"
] |
1 Jun 2023 (<a href="https://arxiv.org/abs/2306.00469v1">v1</a>), last revised 3 Dec 2023 (this version, v2)
|
20 pages
|
Computation (stat.CO)
|
Computation (stat.CO)
|
https://doi.org/10.48550/arXiv.2306.00469
|
This paper investigates the efficient solution of penalized quadratic regressions in high-dimensional settings. A novel and efficient algorithm for ridge-penalized quadratic regression is proposed, leveraging the matrix structures of the regression with interactions. Additionally, an alternating direction method of multipliers (ADMM) framework is developed for penalized quadratic regression with general penalties, including both single and hybrid penalty functions. The approach simplifies the calculations to basic matrix-based operations, making it appealing in terms of both memory storage and computational complexity for solving penalized quadratic regressions in high-dimensional settings.
|
stat-05.zip/2306.00469.pdf
|
2305.08204
|
glmmPen: High Dimensional Penalized Generalized Linear Mixed Models
|
[
"Hillary M. Heiling",
"Naim U. Rashid",
"Quefeng Li",
"Joseph G. Ibrahim"
] |
14 May 2023 (<a href="https://arxiv.org/abs/2305.08204v1">v1</a>), last revised 16 Apr 2024 (this version, v2)
|
Computation (stat.CO)
|
Computation (stat.CO); Methodology (stat.ME)
|
https://doi.org/10.48550/arXiv.2305.08204
|
Generalized linear mixed models (GLMMs) are widely used in research for their ability to model correlated outcomes with non-Gaussian conditional distributions. The proper selection of fixed and random effects is a critical part of the modeling process since model misspecification may lead to significant bias. However, the joint selection of fixed and random effects has historically been limited to lower-dimensional GLMMs, largely due to the use of criterion-based model selection strategies. Here we present the R package glmmPen, one of the first to select fixed and random effects in higher dimension using a penalized GLMM modeling framework. Model parameters are estimated using a Monte Carlo Expectation Conditional Minimization (MCECM) algorithm, which leverages Stan and RcppArmadillo for increased computational efficiency. Our package supports the Binomial, Gaussian, and Poisson families and multiple penalty functions. In this manuscript we discuss the modeling procedure, estimation scheme, and software implementation through application to a pancreatic cancer subtyping study. Simulation results show our method has good performance in selecting both the fixed and random effects in high dimensional GLMMs.
|
stat-05.zip/2305.08204.pdf
|
|
2305.14972
|
Generative AI for Bayesian Computation
|
[
"Nicholas G. Polson",
"Vadim Sokolov"
] |
24 May 2023 (<a href="https://arxiv.org/abs/2305.14972v1">v1</a>), last revised 24 Feb 2024 (this version, v3)
|
Computation (stat.CO)
|
Computation (stat.CO)
|
https://doi.org/10.48550/arXiv.2305.14972
|
Bayesian Generative AI (BayesGen-AI) methods are developed and applied to Bayesian computation. BayesGen-AI reconstructs the posterior distribution by directly modeling the parameter of interest as a mapping (a.k.a. deep learner) from a large simulated dataset. This provides a generator that we can evaluate at the observed data and provide draws from the posterior distribution. This method applies to all forms of Bayesian inference including parametric models, likelihood-free models, prediction and maximum expected utility problems. Bayesian computation is then equivalent to high dimensional non-parametric regression. Bayes Gen-AI main advantage is that it is density-free and therefore provides an alternative to Markov Chain Monte Carlo. It has a number of advantages over vanilla generative adversarial networks (GAN) and approximate Bayesian computation (ABC) methods due to the fact that the generator is simpler to learn than a GAN architecture and is more flexible than kernel smoothing implicit in ABC methods. Design of the Network Architecture requires careful selection of features (a.k.a. dimension reduction) and nonlinear architecture for inference. As a generic architecture, we propose a deep quantile neural network and a uniform base distribution at which to evaluate the generator. To illustrate our methodology, we provide two real data examples, the first in traffic flow prediction and the second in building a surrogate for satellite drag data-set. Finally, we conclude with directions for future research.
|
stat-05.zip/2305.14972.pdf
|
|
2305.10068
|
Stein $Π$-Importance Sampling
|
[
"Congye Wang",
"Wilson Chen",
"Heishiro Kanagawa",
"Chris. J. Oates"
] |
17 May 2023
|
Computation (stat.CO)
|
Computation (stat.CO)
|
https://doi.org/10.48550/arXiv.2305.10068
|
Stein discrepancies have emerged as a powerful tool for retrospective improvement of Markov chain Monte Carlo output. However, the question of how to design Markov chains that are well-suited to such post-processing has yet to be addressed. This paper studies Stein importance sampling, in which weights are assigned to the states visited by a $\Pi$-invariant Markov chain to obtain a consistent approximation of $P$, the intended target. Surprisingly, the optimal choice of $\Pi$ is not identical to the target $P$; we therefore propose an explicit construction for $\Pi$ based on a novel variational argument. Explicit conditions for convergence of Stein $\Pi$-Importance Sampling are established. For $\approx 70\%$ of tasks in the PosteriorDB benchmark, a significant improvement over the analogous post-processing of $P$-invariant Markov chains is reported.
|
stat-05.zip/2305.10068.pdf
|
|
2305.14746
|
Wasserstein Gaussianization and Efficient Variational Bayes for Robust Bayesian Synthetic Likelihood
|
[
"Nhat-Minh Nguyen",
"Minh-Ngoc Tran",
"Christopher Drovandi",
"David Nott"
] |
24 May 2023 (<a href="https://arxiv.org/abs/2305.14746v1">v1</a>), last revised 15 Aug 2024 (this version, v2)
|
Computation (stat.CO)
|
Computation (stat.CO); Machine Learning (stat.ML)
|
https://doi.org/10.48550/arXiv.2305.14746
|
The Bayesian Synthetic Likelihood (BSL) method is a widely-used tool for likelihood-free Bayesian inference. This method assumes that some summary statistics are normally distributed, which can be incorrect in many applications. We propose a transformation, called the Wasserstein Gaussianization transformation, that uses a Wasserstein gradient flow to approximately transform the distribution of the summary statistics into a Gaussian distribution. BSL also implicitly requires compatibility between simulated summary statistics under the working model and the observed summary statistics. A robust BSL variant which achieves this has been developed in the recent literature. We combine the Wasserstein Gaussianization transformation with robust BSL, and an efficient Variational Bayes procedure for posterior approximation, to develop a highly efficient and reliable approximate Bayesian inference method for likelihood-free problems.
|
stat-05.zip/2305.14746.pdf
|
|
2303.02756
|
A New Class of Realistic Spatio-Temporal Processes with Advection and Their Simulation
|
[
"Maria Laura Battagliola",
"Sofia Charlotta Olhede"
] |
5 Mar 2023 (<a href="https://arxiv.org/abs/2303.02756v1">v1</a>), last revised 29 Apr 2024 (this version, v2)
|
Computation (stat.CO)
|
Computation (stat.CO); Methodology (stat.ME)
|
https://doi.org/10.48550/arXiv.2303.02756
|
Traveling phenomena, frequently observed in a variety of scientific disciplines including atmospheric science, seismography, and oceanography, have long suffered from limitations due to lack of realistic statistical modeling tools and simulation methods. Our work primarily addresses this, introducing more realistic and flexible models for spatio-temporal random fields. We break away from the traditional confines of the classic frozen field by either relaxing the assumption of a single deterministic velocity or rethinking the hypothesis regarding the spectrum shape, thus enhancing the realism of our models. While the proposed models stand out for their realism and flexibility, they are also paired with simulation algorithms that are equally or less computationally complex than the commonly used circulant embedding for Gaussian random fields in $\mathbb{R}^{2+1}$. This combination of realistic modeling with efficient simulation methods creates an effective solution for better understanding traveling phenomena.
|
stat-05.zip/2303.02756.pdf
|
|
2303.06422
|
An approximate control variates approach to multifidelity distribution estimation
|
[
"Ruijian Han",
"Boris Kramer",
"Dongjin Lee",
"Akil Narayan",
"Yiming Xu"
] |
11 Mar 2023 (<a href="https://arxiv.org/abs/2303.06422v1">v1</a>), last revised 5 Jul 2023 (this version, v2)
|
41 pages, added additional numerical experiments
|
Computation (stat.CO)
|
Computation (stat.CO); Numerical Analysis (math.NA); Applications (stat.AP)
|
https://doi.org/10.48550/arXiv.2303.06422
|
Forward simulation-based uncertainty quantification that studies the distribution of quantities of interest (QoI) is a crucial component for computationally robust engineering design and prediction. There is a large body of literature devoted to accurately assessing statistics of QoIs, and in particular, multilevel or multifidelity approaches are known to be effective, leveraging cost-accuracy tradeoffs between a given ensemble of models. However, effective algorithms that can estimate the full distribution of QoIs are still under active development. In this paper, we introduce a general multifidelity framework for estimating the cumulative distribution function (CDF) of a vector-valued QoI associated with a high-fidelity model under a budget constraint. Given a family of appropriate control variates obtained from lower-fidelity surrogates, our framework involves identifying the most cost-effective model subset and then using it to build an approximate control variates estimator for the target CDF. We instantiate the framework by constructing a family of control variates using intermediate linear approximators and rigorously analyze the corresponding algorithm. Our analysis reveals that the resulting CDF estimator is uniformly consistent and asymptotically optimal as the budget tends to infinity, with only mild moment and regularity assumptions on the joint distribution of QoIs. The approach provides a robust multifidelity CDF estimator that is adaptive to the available budget, does not require \textit{a priori} knowledge of cross-model statistics or model hierarchy, and applies to multiple dimensions. We demonstrate the efficiency and robustness of the approach using test examples of parametric PDEs and stochastic differential equations including both academic instances and more challenging engineering problems.
|
stat-05.zip/2303.06422.pdf
|
2305.07481
|
Extended ADMM for general penalized quantile regression with linear constraints in big data
|
[
"Yongxin Liu",
"Peng Zeng"
] |
12 May 2023
|
Computation (stat.CO)
|
Computation (stat.CO)
|
https://doi.org/10.48550/arXiv.2305.07481
|
Quantile regression (QR) can be used to describe the comprehensive relationship between a response and predictors. Prior domain knowledge and assumptions in application are usually formulated as constraints of parameters to improve the estimation efficiency. This paper develops methods based on multi-block ADMM to fit general penalized QR with linear constraints of regression coefficients. Different formulations to handle the linear constraints and general penalty are explored and compared. The most efficient one has explicit expressions for each parameter and avoids nested-loop iterations in some existing algorithms. Additionally, parallel ADMM algorithm for big data is also developed when data are stored in a distributed fashion. The stopping criterion and convergence of the algorithm are established. Extensive numerical experiments and a real data example demonstrate the computational efficiency of the proposed algorithms. The details of theoretical proofs and different algorithm variations are presented in Appendix.
|
stat-05.zip/2305.07481.pdf
|
|
2305.17573
|
CRAN Task Views: The Next Generation
|
[
"Achim Zeileis",
"Roger Bivand",
"Dirk Eddelbuettel",
"Kurt Hornik",
"Nathalie Vialaneix"
] |
27 May 2023
|
8 pages, 1 figure
|
Computation (stat.CO)
|
Computation (stat.CO)
|
https://doi.org/10.48550/arXiv.2305.17573
|
CRAN Task Views have been available on the Comprehensive R Archive Network since 2005. They provide guidance about which CRAN packages are relevant for tasks related to a certain topic, and can also facilitate automatic installation of all corresponding packages. Motivated by challenges from the growth of CRAN and the R community as a whole since 2005, all of the task views infrastructure and workflows were rethought and relaunched in 2021/22 in order to facilitate maintenance and to foster deeper interactions with the R community. The redesign encompasses the establishment of a group of CRAN Task View Editors, moving all task view sources to dedicated GitHub repositories, adopting well-documented workflows with a code of conduct, and leveraging R/Markdown files (rather than XML) for the content of the task views.
|
stat-05.zip/2305.17573.pdf
|
2303.05040
|
Modeling metallic fatigue data using the Birnbaum--Saunders distribution
|
[
"Zaid Sawlan",
"Marco Scavino",
"Raul Tempone"
] |
9 Mar 2023 (<a href="https://arxiv.org/abs/2303.05040v1">v1</a>), last revised 6 Jul 2023 (this version, v3)
|
Computation (stat.CO)
|
Computation (stat.CO); Applications (stat.AP)
|
https://doi.org/10.48550/arXiv.2303.05040
|
This work employs the Birnbaum--Saunders distribution to model the fatigue life of metallic materials under cyclic loading and compares it with the normal distribution. Fatigue-limit models are fitted to three datasets of unnotched specimens of 75S-T6 aluminum alloys and carbon laminate with different loading types. A new equivalent stress definition that accounts for the effect of the experiment type is proposed. The results show that the Birnbaum--Saunders distribution consistently outperforms the normal distribution in fitting the fatigue data and provides more accurate predictions of fatigue life and survival probability.
|
stat-05.zip/2303.05040.pdf
|
|
2305.07276
|
multilevLCA: An R Package for Single-Level and Multilevel Latent Class Analysis with Covariates
|
[
"Johan Lyrvall",
"Roberto Di Mari",
"Zsuzsa Bakk",
"Jennifer Oser",
"Jouni Kuha"
] |
12 May 2023 (<a href="https://arxiv.org/abs/2305.07276v1">v1</a>), last revised 10 Apr 2024 (this version, v2)
|
Computation (stat.CO)
|
Computation (stat.CO); Methodology (stat.ME)
|
https://doi.org/10.48550/arXiv.2305.07276
|
This contribution presents a guide to the R package multilevLCA, which offers a complete and innovative set of technical tools for the latent class analysis of single-level and multilevel categorical data. We describe the available model specifications, mainly falling within the fixed-effect or random-effect approaches. Maximum likelihood estimation of the model parameters, enhanced by a refined initialization strategy, is implemented either simultaneously, i.e., in one-step, or by means of the more advantageous two-step estimator. The package features i) semi-automatic model selection when a priori information on the number of classes is lacking, ii) predictors of class membership, and iii) output visualization tools for any of the available model specifications. All functionalities are illustrated by means of a real application on citizenship norms data, which are available in the package.
|
stat-05.zip/2305.07276.pdf
|
|
2303.03576
|
A Survey of Numerical Algorithms that can Solve the Lasso Problems
|
[
"Yujie Zhao",
"Xiaoming Huo"
] |
7 Mar 2023
|
Computation (stat.CO)
|
Computation (stat.CO); Machine Learning (stat.ML)
|
https://doi.org/10.48550/arXiv.2303.03576
|
In statistics, the least absolute shrinkage and selection operator (Lasso) is a regression method that performs both variable selection and regularization. There is a lot of literature available, discussing the statistical properties of the regression coefficients estimated by the Lasso method. However, there lacks a comprehensive review discussing the algorithms to solve the optimization problem in Lasso. In this review, we summarize five representative algorithms to optimize the objective function in Lasso, including the iterative shrinkage threshold algorithm (ISTA), fast iterative shrinkage-thresholding algorithms (FISTA), coordinate gradient descent algorithm (CGDA), smooth L1 algorithm (SLA), and path following algorithm (PFA). Additionally, we also compare their convergence rate, as well as their potential strengths and weakness.
|
stat-05.zip/2303.03576.pdf
|
|
2303.06398
|
Variational Gaussian filtering via Wasserstein gradient flows
|
[
"Adrien Corenflos",
"Hany Abdulsamad"
] |
11 Mar 2023 (<a href="https://arxiv.org/abs/2303.06398v1">v1</a>), last revised 19 Jun 2023 (this version, v2)
|
5 pages, 2 figures, double column, minor modifications compared to version 1 (more experiments + typos). Accepted as a conference paper to EUSIPCO 2023
|
Computation (stat.CO)
|
Computation (stat.CO); Computational Engineering, Finance, and Science (cs.CE); Systems and Control (eess.SY); Machine Learning (stat.ML)
|
https://doi.org/10.48550/arXiv.2303.06398
|
We present a novel approach to approximate Gaussian and mixture-of-Gaussians filtering. Our method relies on a variational approximation via a gradient-flow representation. The gradient flow is derived from a Kullback--Leibler discrepancy minimization on the space of probability distributions equipped with the Wasserstein metric. We outline the general method and show its competitiveness in posterior representation and parameter estimation on two state-space models for which Gaussian approximations typically fail: systems with multiplicative noise and multi-modal state distributions.
|
stat-05.zip/2303.06398.pdf
|
2303.05992
|
Statistical optimization of expensive multi-response black-box functions
|
[
"Andreas Mändle",
"Werner Brannath",
"Werner Wosniok"
] |
10 Mar 2023
|
29 pages, 26 figures, 1 table
|
Computation (stat.CO)
|
Computation (stat.CO); Optimization and Control (math.OC); Applications (stat.AP); Methodology (stat.ME)
|
https://doi.org/10.48550/arXiv.2303.05992
|
Assume that a set of $P$ process parameters $p_i$, $i=1,\dots,P$, determines the outcome of a set of $D$ descriptor variables $d_j$, $j=1,\dots,D$, via an unknown functional relationship $\phi: \mathbf{p} \mapsto \mathbf{d}, \, \mathbb{R}^{P} \to \mathbb{R}^{D}$, where $\mathbf{p}=(p_1,\dots,p_{P})$, $\mathbf{d}=(d_1,\dots,d_{D})$. It is desired to find appropriate values $\mathbf{\hat p} = ({\hat p}_1,\dots, {\hat p}_P)$ for the process parameters such that the corresponding values of the descriptor variables $\phi (\mathbf {\hat p})$ are close to a given target $\mathbf d^*=(d^*_1,\dots,d^*_D)$, assuming that at least one exact solution exists. A sequential approach using dimension reduction techniques has been developed to achieve this. In a simulation study, results of the suggested approach and the algorithms NSGA-II, SMS-EMOA and MOEA/D are compared.
|
stat-05.zip/2303.05992.pdf
|
2306.14976
|
General adjoint-differentiated Laplace approximation
|
[
"Charles C. Margossian"
] |
26 Jun 2023
|
Computation (stat.CO)
|
Computation (stat.CO)
|
https://doi.org/10.48550/arXiv.2306.14976
|
The hierarchical prior used in Latent Gaussian models (LGMs) induces a posterior geometry prone to frustrate inference algorithms. Marginalizing out the latent Gaussian variable using an integrated Laplace approximation removes the offending geometry, allowing us to do efficient inference on the hyperparameters. To use gradient-based inference we need to compute the approximate marginal likelihood and its gradient. The adjoint-differentiated Laplace approximation differentiates the marginal likelihood and scales well with the dimension of the hyperparameters. While this method can be applied to LGMs with any prior covariance, it only works for likelihoods with a diagonal Hessian. Furthermore, the algorithm requires methods which compute the first three derivatives of the likelihood with current implementations relying on analytical derivatives. I propose a generalization which is applicable to a broader class of likelihoods and does not require analytical derivatives of the likelihood. Numerical experiments suggest the added flexibility comes at no computational cost: on a standard LGM, the new method is in fact slightly faster than the existing adjoint-differentiated Laplace approximation. I also apply the general method to an LGM with an unconventional likelihood. This example highlights the algorithm's potential, as well as persistent challenges.
|
stat-05.zip/2306.14976.pdf
|
|
2307.01111
|
Nonparametric Bayesian approach for quantifying the conditional uncertainty of input parameters in chained numerical models
|
[
"Oumar Baldé",
"Guillaume Damblin",
"Amandine Marrel",
"Antoine Bouloré",
"Loïc Giraldi"
] |
3 Jul 2023 (<a href="https://arxiv.org/abs/2307.01111v1">v1</a>), last revised 2 Oct 2023 (this version, v2)
|
39 pages, 12 figures
|
Computation (stat.CO)
|
Computation (stat.CO); Methodology (stat.ME)
|
https://doi.org/10.48550/arXiv.2307.01111
|
Nowadays, numerical models are widely used in most of engineering fields to simulate the behaviour of complex systems, such as for example power plants or wind turbine in the energy sector. Those models are nevertheless affected by uncertainty of different nature (numerical, epistemic) which can affect the reliability of their predictions. We develop here a new method for quantifying conditional parameter uncertainty within a chain of two numerical models in the context of multiphysics simulation. More precisely, we aim to calibrate the parameters $\theta$ of the second model of the chain conditionally on the value of parameters $\lambda$ of the first model, while assuming the probability distribution of $\lambda$ is known. This conditional calibration is carried out from the available experimental data of the second model. In doing so, we aim to quantify as well as possible the impact of the uncertainty of $\lambda$ on the uncertainty of $\theta$. To perform this conditional calibration, we set out a nonparametric Bayesian formalism to estimate the functional dependence between $\theta$ and $\lambda$, denoted by $\theta(\lambda)$. First, each component of $\theta(\lambda)$ is assumed to be the realization of a Gaussian process prior. Then, if the second model is written as a linear function of $\theta(\lambda)$, the Bayesian machinery allows us to compute analytically the posterior predictive distribution of $\theta(\lambda)$ for any set of realizations $\lambda$. The effectiveness of the proposed method is illustrated on several analytical examples.
|
stat-05.zip/2307.01111.pdf
|
2307.11648
|
Sparse inverse Cholesky factorization of dense kernel matrices by greedy conditional selection
|
[
"Stephen Huan",
"Joseph Guinness",
"Matthias Katzfuss",
"Houman Owhadi",
"Florian Schäfer"
] |
21 Jul 2023 (<a href="https://arxiv.org/abs/2307.11648v1">v1</a>), last revised 9 May 2025 (this version, v2)
|
Accepted for publication in SIAM JUQ
|
Computation (stat.CO)
|
Computation (stat.CO); Numerical Analysis (math.NA)
|
https://doi.org/10.48550/arXiv.2307.11648
|
Dense kernel matrices resulting from pairwise evaluations of a kernel function arise naturally in machine learning and statistics. Previous work in constructing sparse approximate inverse Cholesky factors of such matrices by minimizing Kullback-Leibler divergence recovers the Vecchia approximation for Gaussian processes. These methods rely only on the geometry of the evaluation points to construct the sparsity pattern. In this work, we instead construct the sparsity pattern by leveraging a greedy selection algorithm that maximizes mutual information with target points, conditional on all points previously selected. For selecting $k$ points out of $N$, the naive time complexity is $\mathcal{O}(N k^4)$, but by maintaining a partial Cholesky factor we reduce this to $\mathcal{O}(N k^2)$. Furthermore, for multiple ($m$) targets we achieve a time complexity of $\mathcal{O}(N k^2 + N m^2 + m^3)$, which is maintained in the setting of aggregated Cholesky factorization where a selected point need not condition every target. We apply the selection algorithm to image classification and recovery of sparse Cholesky factors. By minimizing Kullback-Leibler divergence, we apply the algorithm to Cholesky factorization, Gaussian process regression, and preconditioning with the conjugate gradient, improving over $k$-nearest neighbors selection.
|
stat-05.zip/2307.11648.pdf
|
2307.12438
|
Multifidelity Covariance Estimation via Regression on the Manifold of Symmetric Positive Definite Matrices
|
[
"Aimee Maurais",
"Terrence Alsup",
"Benjamin Peherstorfer",
"Youssef Marzouk"
] |
23 Jul 2023 (<a href="https://arxiv.org/abs/2307.12438v1">v1</a>), last revised 5 Sep 2024 (this version, v3)
|
To appear in the SIAM Journal on Mathematics of Data Science (SIMODS)
|
Computation (stat.CO)
|
Computation (stat.CO); Machine Learning (cs.LG); Numerical Analysis (math.NA)
|
https://doi.org/10.48550/arXiv.2307.12438
|
We introduce a multifidelity estimator of covariance matrices formulated as the solution to a regression problem on the manifold of symmetric positive definite matrices. The estimator is positive definite by construction, and the Mahalanobis distance minimized to obtain it possesses properties enabling practical computation. We show that our manifold regression multifidelity (MRMF) covariance estimator is a maximum likelihood estimator under a certain error model on manifold tangent space. More broadly, we show that our Riemannian regression framework encompasses existing multifidelity covariance estimators constructed from control variates. We demonstrate via numerical examples that the MRMF estimator can provide significant decreases, up to one order of magnitude, in squared estimation error relative to both single-fidelity and other multifidelity covariance estimators. Furthermore, preservation of positive definiteness ensures that our estimator is compatible with downstream tasks, such as data assimilation and metric learning, in which this property is essential.
|
stat-05.zip/2307.12438.pdf
|
2307.02428
|
Sampling lattice points in a polytope: a Bayesian biased algorithm with random updates
|
[
"Miles Bakenhus",
"Sonja Petrović"
] |
5 Jul 2023
|
22 pages, 12 figures
|
Computation (stat.CO)
|
Computation (stat.CO); Commutative Algebra (math.AC); Statistics Theory (math.ST)
|
https://doi.org/10.48550/arXiv.2307.02428
|
The set of nonnegative integer lattice points in a polytope, also known as the fiber of a linear map, makes an appearance in several applications including optimization and statistics. We address the problem of sampling from this set using three ingredients: an easy-to-compute lattice basis of the constraint matrix, a biased sampling algorithm with a Bayesian framework, and a step-wise selection method. The bias embedded in our algorithm updates sampler parameters to improve fiber discovery rate at each step chosen from previously discovered elements. We showcase the performance of the algorithm on several examples, including fibers that are out of reach for the state-of-the-art Markov bases samplers.
|
stat-05.zip/2307.02428.pdf
|
2306.17776
|
A multivariate heavy-tailed integer-valued GARCH process with EM algorithm-based inference
|
[
"Yuhyeong Jang",
"Raanju R. Sundararajan",
"Wagner Barreto-Souza"
] |
30 Jun 2023
|
32pages, 14figures
|
Computation (stat.CO)
|
Computation (stat.CO)
|
https://doi.org/10.48550/arXiv.2306.17776
|
A new multivariate integer-valued Generalized AutoRegressive Conditional Heteroscedastic process based on a multivariate Poisson generalized inverse Gaussian distribution is proposed. The estimation of parameters of the proposed multivariate heavy-tailed count time series model via maximum likelihood method is challenging since the likelihood function involves a Bessel function that depends on the multivariate counts and its dimension. As a consequence, numerical instability is often experienced in optimization procedures. To overcome this computational problem, two feasible variants of the Expectation-Maximization (EM) algorithm are proposed for estimating parameters of our model under low and high-dimensional settings. These EM algorithm variants provide computational benefits and help avoid the difficult direct optimization of the likelihood function from the proposed model. Our model and proposed estimation procedures can handle multiple features such as modeling of multivariate counts, heavy-taildness, overdispersion, accommodation of outliers, allowances for both positive and negative autocorrelations, estimation of cross/contemporaneous-correlation, and the efficient estimation of parameters from both statistical and computational points of view. Extensive Monte Carlo simulation studies are presented to assess the performance of the proposed EM algorithms. An application to modeling bivariate count time series data on cannabis possession-related offenses in Australia is discussed.
|
stat-05.zip/2306.17776.pdf
|
2307.00219
|
Iterative conditional replacement algorithm for conditionally specified models
|
[
"Kun-Lin Kuo",
"Yuchung J. Wang"
] |
1 Jul 2023
|
21 pages, 3 figures
|
Computation (stat.CO)
|
Computation (stat.CO)
|
https://doi.org/10.48550/arXiv.2307.00219
|
The sample-based Gibbs sampler has been the dominant method for approximating joint distribution from a collection of compatible full-conditional distributions. However for conditionally specified model, mixtures of incompatible full and non-full conditional distributions are the realities; but, their updating orders are hard to identified. We propose a new algorithm, the Iterative Conditional Replacement (ICR), that produces distributional approximations toward the stationary distributions, dispensing Markov chain entirely. ICR always converges, and it produces mutually stationary distributions, which will be consistent among one another when the conditional distributions are compatible. Examples show ICR to be superior in quality, while being more parallelizable and requiring little effort in monitoring its convergence. Last, we propose an ensemble approach to decide the final model.
|
stat-05.zip/2307.00219.pdf
|
Subsets and Splits
CS Papers Since 2025
The query filters and organizes computer science papers submitted on or after January 1, 2025, providing a structured view of their metadata, which is useful for exploring recent trends in computer science research.
CS Papers from 2013 Onwards
Retrieves detailed information about computer science papers submitted after 2013, providing insights into the volume and subjects of research over time.
Quantitative Finance Papers Query
Retrieves recent papers related to quantitative finance and trading from arXiv, providing insights into recent research trends in these fields.
Quant Finance Papers
Retrieves the latest 100 papers related to quantitative finance, trading, or algorithmic trading from the arXiv dataset, providing titles, authors, submission dates, and abstracts.
Recent ArXiv Papers Filtered
Filters and retrieves papers submitted after January 1, 2025, with submission dates in a specific format, providing a basic overview of recent submissions.
ArXiv Papers by Year 202
The query filters records based on a specific date format and year range, providing a basic subset of the data for further analysis.
Papers Submitted in 2025
Retrieves all records from the train dataset where the submission date is in the year 2025, providing a basic filter on the data.