Elsevier, Statistics and Probability Letters, 16(78), p. 2744-2750, 2008
DOI: 10.1016/j.spl.2008.03.022
Full text: Download
The problem of quantile estimation in general semiparametric regression models is considered. We derive plug-in kernel-based estimators, investigate their asymptotic distribution and establish the semiparametric efficiency of these estimators under mild assumptions. We apply our methodology in an example in nutritional epidemiology. The generalization to the important case where responses are missing at random is also addressed.