A0727
Title: Zero-inflated Poisson models with measurement error in the response
Authors: Qihuang Zhang - McGill University (Canada) [presenting]
Abstract: Zero-inflated count data arise frequently from genomics studies. Analysis of such data is often based on a mixture model, which facilitates excess zeros in combination with a Poisson distribution, and various inference methods have been proposed under such a model. Those analysis procedures, however, are challenged by the presence of measurement errors in count responses. A new measurement error model is proposed to describe error-contaminated count data. It is shown that ignoring the measurement error effects in the analysis may generally lead to invalid inference results, and meanwhile, situations are identified where ignoring measurement error can still yield consistent estimators. Furthermore, a Bayesian method is proposed to address the effects of measurement error under the zero-inflated Poisson model and discuss the identifiability issues. A data-augmentation algorithm is developed that is easy to implement. Simulation studies are conducted to evaluate the performance of the proposed method. The method is applied to analyze the data arising from a prostate adenocarcinoma genomic study.
