Testing goodness-of-fit and conditional independence with approximate co-sufficient sampling

TL;DR

This paper introduces approximate co-sufficient sampling (aCSS), a generalized goodness-of-fit testing method that extends CSS to a wide range of parametric models using approximate sufficiency, ensuring valid testing with controlled Type I error.

Contribution

The paper develops aCSS, extending CSS to models with asymptotically-efficient estimators, and provides theoretical and empirical analysis of its error control and power.

Findings

aCSS controls Type I error asymptotically

The method is applicable to a broad class of parametric models

Simulation results demonstrate effective power and error control

Abstract

Goodness-of-fit (GoF) testing is ubiquitous in statistics, with direct ties to model selection, confidence interval construction, conditional independence testing, and multiple testing, just to name a few applications. While testing the GoF of a simple (point) null hypothesis provides an analyst great flexibility in the choice of test statistic while still ensuring validity, most GoF tests for composite null hypotheses are far more constrained, as the test statistic must have a tractable distribution over the entire null model space. A notable exception is co-sufficient sampling (CSS): resampling the data conditional on a sufficient statistic for the null model guarantees valid GoF testing using any test statistic the analyst chooses. But CSS testing requires the null model to have a compact (in an information-theoretic sense) sufficient statistic, which only holds for a very limited class of models; even for a null model as simple as logistic regression, CSS testing is powerless. In this paper, we leverage the concept of approximate sufficiency to generalize CSS testing to essentially any parametric model with an asymptotically-efficient estimator; we call our extension "approximate CSS" (aCSS) testing. We quantify the finite-sample Type I error inflation of aCSS testing and show that it is vanishing under standard maximum likelihood asymptotics, for any choice of test statistic. We apply our proposed procedure both theoretically and in simulation to a number of models of interest to demonstrate its finite-sample Type I error and power.