Binary regression with stochastic covariates


Oral E.

COMMUNICATIONS IN STATISTICS-THEORY AND METHODS, vol.35, no.8, pp.1429-1447, 2006 (Peer-Reviewed Journal) identifier identifier

  • Publication Type: Article / Article
  • Volume: 35 Issue: 8
  • Publication Date: 2006
  • Doi Number: 10.1080/03610920600637123
  • Journal Name: COMMUNICATIONS IN STATISTICS-THEORY AND METHODS
  • Journal Indexes: Science Citation Index Expanded, Scopus
  • Page Numbers: pp.1429-1447
  • Keywords: binary data, logistic regression, modified maximum likelihood, non-normality, order statistics, stochastic covariate, MAXIMUM-LIKELIHOOD, NONNORMAL REGRESSION, DISTRIBUTIONS, PARAMETERS, MODELS

Abstract

In binary regression the risk factor X has been treated in the literature as a non-stochastic variable. In most situations, however, X is stochastic. We present solutions applicable to such situations. We show that our solutions are more precise than those obtained by treating X as non-stochastic when, in fact, it is stochastic.