634 – Challenges and Advances in Evaluation of Clinical Trials
Small Area Prediction of the Mean of a Binomial Random Variable
Andreea Erciulescu
Iowa State University
Wayne Fuller
Iowa State University
Direct estimates for small areas or subpopulations may not be reliable because of small sample sizes for such objects. Procedures based on implicit or explicit models have been used to construct better estimates for given small areas, by exploiting auxiliary information. In this paper we consider binary responses, and investigate different cases associated with different amounts of available information. We use generalized linear mixed models and present bias and mean squared error results for different prediction methods.