English

Pseudo-$R^2$ statistics under complex sampling

Methodology 2017-01-27 v1

Abstract

Model summaries based on the ratio of fitted and null likelihoods have been proposed for generalised linear models, reducing to the familiar R2R^2 coefficient of determination in the Gaussian model with identity link. In this note I show how to define the Cox--Snell and Nagelkerke summaries under arbitrary probability sampling designs, giving a design-consistent estimator of the population model summary. I also show that for logistic regression models under case--control sampling the usual Cox--Snell and Nagelkerke R2R^2 are not design-consistent, but are systematically larger than would be obtained with a cross-sectional or cohort sample, even in settings where the weighted and unweighted logistic regression estimators are similar or identical.

Keywords

Cite

@article{arxiv.1701.07745,
  title  = {Pseudo-$R^2$ statistics under complex sampling},
  author = {Thomas Lumley},
  journal= {arXiv preprint arXiv:1701.07745},
  year   = {2017}
}
R2 v1 2026-06-22T18:01:29.709Z