arXiv subjects

Elizabeth Bersson

Publications and source records attributed to Elizabeth Bersson.

5 recordsLinked to original sources

arXiv2026

The Curious Case of the Default Settings: Evaluating Default Performance of Variational Inference Software

We systematically evaluate a suite of off-the-shelf variational inference (VI) software packages from the perspective of a standard practitioner. Using simple analytic benchmark models, we assess the accuracy and stability of the default VI settings in PyMC, NumPyro, and TensorFlow Probability. Unlike previous research focusing on methodological advances, our evaluation emphasizes software implementation and the default configurations that typical users encounter. Our results show that default settings can yield biased approximations of posterior summaries even for simple one-dimensional conjugate models, controls of initialization and transformations differ between software implementations, and relying on defaults may yield silent failures or poor approximations.

stat.CO↗

arXiv2025

Feature aware covariance estimation, with application to mixtures of chemical exposures

The motivation of this article is to improve inferences on the covariation in environmental exposures, motivated by data from a study of Toddlers Exposure to SVOCs in Indoor Environments (TESIE). The challenge is that the sample size is limited, so empirical covariance provides a poor estimate. In related applications, Bayesian factor models have been popular; these approaches express the covariance as low rank plus diagonal and can infer the number of factors adaptively. However, they have the disadvantage of shrinking towards a diagonal covariance, often under estimating important covariation patterns in the data. Alternatively, the dimensionality problem is addressed by collapsing the detailed exposure data within chemical classes, potentially obscuring important information. We apply a feature aware covariance regression extension of Bayesian factor analysis, which improves performance by including information from features summarizing properties of the different exposures. This approach enables shrinkage to more flexible covariance structures, reducing the over-shrinkage problem, as we illustrate in the TESIE data using various chemical features.

stat.ME↗

arXiv2023

Frequentist Prediction Sets for Species Abundance using Indirect Information

Citizen science databases that consist of volunteer-led sampling efforts of species communities are relied on as essential sources of data in ecology. Summarizing such data across counties with frequentist-valid prediction sets for each county provides an interpretable comparison across counties of varying size or composition. As citizen science data often feature unequal sampling efforts across a spatial domain, prediction sets constructed with indirect methods that share information across counties may be used to improve precision. In this article, we present a nonparametric framework to obtain precise prediction sets for a multinomial random sample based on indirect information that maintain frequentist coverage guarantees for each county. We detail a simple algorithm to obtain prediction sets for each county using indirect information where the computation time does not depend on the sample size and scales nicely with the number of species considered. The indirect information may be estimated by a proposed empirical Bayes procedure based on information from auxiliary data. Our approach makes inference for under-sampled counties more precise, while maintaining area-specific frequentist validity for each county. Our method is used to provide a useful description of avian species abundance in North Carolina, USA based on citizen science data from the eBird database.

stat.ME↗

arXiv2023

Bayesian Covariance Estimation for Multi-group Matrix-variate Data

Multi-group covariance estimation for matrix-variate data with small within group sample sizes is a key part of many data analysis tasks in modern applications. To obtain accurate group-specific covariance estimates, shrinkage estimation methods which shrink an unstructured, group-specific covariance either across groups towards a pooled covariance or within each group towards a Kronecker structure have been developed. However, in many applications, it is unclear which approach will result in more accurate covariance estimates. In this article, we present a hierarchical prior distribution which flexibly allows for both types of shrinkage. The prior linearly combines shrinkage across groups towards a shared pooled covariance and shrinkage within groups towards a group-specific Kronecker covariance. We illustrate the utility of the proposed prior in speech recognition and an analysis of chemical exposure data.

stat.ME↗

arXiv2022

Optimal Conformal Prediction for Small Areas

Existing inferential methods for small area data involve a trade-off between maintaining area-level frequentist coverage rates and improving inferential precision via the incorporation of indirect information. In this article, we propose a method to obtain an area-level prediction region for a future observation which mitigates this trade-off. The proposed method takes a conformal prediction approach in which the conformity measure is the posterior predictive density of a working model that incorporates indirect information. The resulting prediction region has guaranteed frequentist coverage regardless of the working model, and, if the working model assumptions are accurate, the region has minimum expected volume compared to other regions with the same coverage rate. When constructed under a normal working model, we prove such a prediction region is an interval and construct an efficient algorithm to obtain the exact interval. We illustrate the performance of our method through simulation studies and an application to EPA radon survey data.

stat.ME↗