COLLOQUIUM ON THE STATISTICAL SCIENCES THE SCHOOL OF STATISTICS University of the Philippines invites all students and faculty to a colloquium on Analysis of autocorrelated count data: A combined estimating function approach to be given by DR. ALEXANDER R. DE LEON Friday, 13 May, 5:00 p.m. UPSS Auditorium ABSTRACT A flexible semi-parametric model for autocorrelated count data is proposed. Unlike earlier models available in the literature, the model does not require construction of a likelihood function and only entails the specification of the first two conditional moments. A simple and flexible combined estimating function approach that makes efficient use of the information contained in the data is adopted for the model in the absence of a likelihood function. Simulation studies are conducted to study the performance of the semi-parametric method when the likelihood is either correctly specified or misspecified. The methodology is illustrated using time series data on monthly polio counts in the United States from January 1970 to December 1987. Keywords: Autoregressive conditional Poisson (ACP), generalized double ACP (GDACP), Godambe information, INGARCH, martingale, stochastic autoregressive mean (SAM) About the author: DR. DE LEON is currently an Associate Professor at the Department of Mathematics and Statistics, University of Calgary, Canada. Dr. De Leon edited the book Analysis of Mixed Data: Methods and Applications published by Chapman and Hall/RCR in 2013