Evaluating Parameter Uncertainty in the Poisson Lognormal Model with Corrected Variational Estimators

TL;DR

This paper assesses the statistical properties of variational estimators in the Poisson-Log-Normal model, deriving a Sandwich estimator for variance, and demonstrates its effectiveness through simulations and real scRNA-seq data analysis.

Contribution

It introduces a theoretical framework for the consistency and variance estimation of PLN model parameters using M-estimation and the Sandwich estimator.

Findings

Sandwich estimator provides accurate variance estimates for PLN parameters.

Simulation studies show improved coverage over Fisher Information method.

Validation on scRNA-seq data confirms practical applicability.

Abstract

Count data analysis is essential across diverse fields, from ecology and accident analysis to single-cell RNA sequencing (scRNA-seq) and metagenomics. While log transformations are computationally efficient, model-based approaches such as the Poisson-Log-Normal (PLN) model provide robust statistical foundations and are more amenable to extensions. The PLN model, with its latent Gaussian structure, not only captures overdispersion but also enables correlation between variables and inclusion of covariates, making it suitable for multivariate count data analysis. Variational approximations are a golden standard to estimate parameters of complex latent variable models such as PLN, maximizing a surrogate likelihood. However, variational estimators lack theoretical statistical properties such as consistency and asymptotic normality. In this paper, we investigate the consistency and variance estimation of PLN parameters using M-estimation theory. We derive the Sandwich estimator, previously studied in Westling and McCormick (2019), specifically for the PLN model. We compare this approach to the variational Fisher Information method, demonstrating the Sandwich estimator's effectiveness in terms of coverage through simulation studies. Finally, we validate our method on a scRNA-seq dataset.