Inference with extremes: Accounting for Extreme Values in Count Regression Models

Abstract:

Processes which occasionally, but not always, produce extreme values are notoriously diicult to model as a small number of extreme observations may have a large impact on the results. Existing methods for handling extreme values are oen arbitrary and leave researchers without guidance regarding this problem. In this paper we propose an Extreme Value and Zero Inflated Negative Binomial (EVZINB) regression model, which allows for separate modeling of extreme and non-extreme observations, to solve this problem. The EVZINB model oers an elegant solution to modeling data with extreme values and allows researchers to draw additional inferences about both extreme and non-extreme observations. We illustrate the usefulness of the EVZINB model by replicating a study on one-sided violence against civilians.

Authors:

Randahl David and Johan Vegelius