Estimation Based on Pooled Data in Human Biomonitoring and Statistical Genetics

Abstract

Pooling is a cost-effective way to collect data. However, estimation is complicated by the often intractable distributions of the observed pool averages. In this thesis, we consider two applications involving pooled data. The first is to use aggregate data collected from pools of individuals to estimate the levels of individual exposure for various environmental biochemicals. We propose a quasi empirical Bayes estimation approach based on a Gaussian working likelihood which enables pooling of information across different demographic groups. The new estimator out-performs an existing estimator in simulation studies. We consider haplotype frequency estimation from pooled genotype data in our second application. A quick collapsed data estimator is proposed which does not lose much efficiency for rare genetic variants. For more efficient estimates, we propose a way to construct a data-based list of possible haplotypes to be used in conjunction with the expectation-maximization (EM) algorithm to make it more feasible computationally.