A logistic regression analysis approach for sample survey data based on phi-divergence measures

摘要

A new family of minimum distance estimators for binary logistic regressionmodels based on $\phi$-divergence measures is introduced. The so called "pseudominimum phi-divergence estimator"(PM$\phi$E) family is presented as anextension of "minimum phi-divergence estimator" (M$\phi$E) for general samplesurvey designs and contains, as a particular case, the pseudo maximumlikelihood estimator (PMLE) considered in Roberts et al. \cite{r}. Through asimulation study it is shown that some PM$\phi$Es have a better behaviour, interms of efficiency, than the PMLE.