Date of Award

1982

Degree Type

Dissertation

Degree Name

Doctor of Philosophy

Abstract

Asymptotic series expansions for the bias and mean square error of the logit estimator are developed. In addition, an estimator of the bias and asymptotic series for the expectation of this estimator and the estimator of the mean square error are derived. These formulations have been expressed in such a way that general coefficients can be calculated and then implemented for any choice of the parameters involved in the logit estimation problem. The formulas developed are applied, providing similar asymptotic series, for the estimation of the log odds ratio.;A general method for the linear combination of estimators is proposed. This procedure is applied to combine log odds ratios and new estimators are developed for the estimation of a common odds ratio. The small sample properties of these estimators and some widely used estimators are examined and compared in a Monte Carlo simulation. Since no one estimator is uniformly the best, a scheme for applying the most appropriate estimator in a given situation is proposed. Illustrations of settings in which this approach is feasible are examined.

Share

COinS
 
 

To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.