Small Sample Properties of Parametric and Nonparametric Estimators in Q

by Dongryoen Park and Sangun Park.

Abstract: In bioassay, the logit model is the most widely used parametric model. However, the exact form of the response curve is usually unknown and even very complicated, so it is likely that the true model does not follow the logit model. Therefore, according to well known asymptotic results, when the sample size is very large, we should probably use nonparametric regression rather than the logit model unless the exact form of the true response curve is known. In practice, however, we can not increase the sample size infinitely, so the asymptotic result would not be so useful. In this paper, we would like to compare the small sample properties of the logit model and the nonparametric estimator. As the nonparametric method, we choose the locally weighted quasi--likelihood estimator. A Monte Carlo study was done under various circumstances, and it turned out that the locally weighted quasi--likelihood estimator is very competitive in the small sample situation.

Key Words: Local quasi--likelihood, Logit model, Nonparametric regression, Response curve, Small sample.

Dongryeon Park,
Sangun Park,

Editor: Avner Bar-Hen,

READING THE ARTICLE: You can read the article in portable document (.pdf) format (146506 bytes.)

NOTE: The content of this article is the intellectual property of the authors, who retains all rights to future publication.

This page has been accessed 2179 times since July 24, 2006.

Return to the InterStat Home Page.