Introducing a Flexible Zero-Inflated Count Model to Address Data Dispersion

• Activity Number: 395 - Recent Advances in Zero-Inflated Regression Models
• Type: Topic Contributed
• Date/Time: Tuesday, August 1, 2017 : 2:00 PM to 3:50 PM
• Sponsor: Biometrics Section
• Author(s): Kimberly F Sellers* and Andrew Raim
• Companies: Georgetown University and U.S. Census Bureau
• Keywords: Conway-Maxwell-Poisson ; Over-dispersion ; Under-dispersion ; Excess Zeroes

Abstract

While excess zeroes result in an increased chance for over-dispersion in count data, the implication is not guaranteed. One should instead consider a flexible distribution that not only can account for excess zeroes, but can also address potential over- or under-dispersion. A zero-inflated Conway-Maxwell-Poisson (ZICMP) regression allows for modeling the relationship between explanatory and response variables, accounting for both excess zeroes and dispersion. This talk introduces the ZICMP model and illustrates its flexibility, highlighting various statistical properties and model fit through several examples.