This Article Full Text (PDF) References Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Saved Citations Download to citation manager Request Permissions Request Reprints Add to My Marked Citations Citing Articles Citing Articles via Google Scholar Citing Articles via Scopus Google Scholar Articles by Böhning, D. Articles by Viwatwongkasem, C. Search for Related Content PubMed PubMed Citation Articles by Böhning, D. Articles by Viwatwongkasem, C. Social Bookmarking                   What's this?

Revisiting proportion estimators

Free University Berlin=Humboldt University at Berlin, Berlin, Germany, boehning{at}zedat.fu-berlin.de

Chukiat Viwatwongkasem

Department of Biostatistics, Faculty of Public Health, Mahidol University, Bangkok, Thailand

Proportion estimators are quite frequently used in many application areas. The conventional proportion estimator (number of events divided by sample size) encounters a number of problems when the data are sparse as will be demonstrated in various settings. The problem of estimating its variance when sample sizes become small is rarely addressed in a satisfying framework. Specifically, we have in mind applications like the weighted risk difference in multicenter trials or stratifying risk ratio estimators (to adjust for potential confounders) in epidemiological studies. It is suggested to estimate p using the parametric family (see PDF for character) and p(1 - p) using (see PDF for character), where (see PDF for character). We investigate the estimation problem of choosing c 0 from various perspectives including minimizing the average mean squared error of (see PDF for character), average bias and average mean squared error of (see PDF for character). The optimal value of c for minimizing the average mean squared error of (see PDF for character) is found to be independent of n and equals c = 1. The optimal value of c for minimizing the average mean squared error of (see PDF for character) is found to be dependent of n with limiting value c = 0.833. This might justifiy to use a near-optimal value of c = 1 in practice which also turns out to be beneficial when constructing confidence intervals of the form (see PDF for character).

Statistical Methods in Medical Research, Vol. 14, No. 2, 147-169 (2005)
DOI: 10.1191/0962280205sm393oa


CiteULike    Connotea    Del.icio.us    Digg    Reddit    Technorati    What's this?