Title:  Statistical Modeling with Zero-inflated Data

Time:  Saturday,  11:00 AM - 12:00 PM

Abstract: Count data for infrequent events often exhibit larges numbers of zero counts. For such "zero-inflated" data, common distributions for count data such as Poisson and negative binomial are not flexible enough to model the distribution. We illustrate the use of the zero inflated Poisson distribution for hospital admissions counts; from frequentist and Bayesian perspectives,  and address related issues such as power/sample size calculations.

Bio: Michael Arciero received his Ph.D in Mathematics from the University of Rhode Island in 2001. He has been at the University of New England since 2004, where he has developed programs in Applied Mathematics, Data Science, and Statistics. His current and recent work modeling health outcomes includes collaborations with Senscio Systems and Maine Medical Center.