Sample Size Determination for Comparing Two Poisson Rates with Underreported Counts |
| |
Authors: | James Stamey Athanassios Katsis |
| |
Institution: | 1. Department of Statistical Science , Baylor University , Waco, Texas, USA james_stamey@baylor.edu;3. Department of Social and Educational Policy , University of Peloponnese , Korinthos, Greece |
| |
Abstract: | The optimal sample size comparing two Poisson rates when the counts are underreported is investigated. We consider two sampling scenarios. We first consider the case where only underreported data will be sampled and rely on informative prior distributions to obtain posterior identifiability. We also consider the case where an expensive infallible search method and a fallible method are available. An interval based sample size criterion is used in both sampling scenarios. Since the posterior distributions of the two rates are functions of confluent hypergeometric and hypergeometric functions simulation based methods are necessary to perform the sample size determination scheme. |
| |
Keywords: | Average length criterion MCMC Underreporting |
|
|