Count Models Based on Weibull Interarrival Times

26 Pages Posted: 30 May 2005

See all articles by Eric Bradlow

Eric Bradlow

University of Pennsylvania - Marketing Department

Peter Fader

University of Pennsylvania - Marketing Department

Moshe Adrian

University of Illinois at Urbana-Champaign

Blakeley B. McShane

Northwestern University - Kellogg School of Management

Date Written: January 2006

Abstract

The widespread popularity and use of both the Poisson and negative binomial models for count data arises, in part, from their derivation as the number of arrivals in a given time period assuming exponenitally distributed interarrival times (without and with heterogeneity in the underlying base rates respectively). However, with that clean theory comes some limitations including limited flexibility in the assumed underlying arrival rate distribution and the inability to model underdispersed counts (variance less than the mean). While extant research has addressed some of these issues, there still remain numerous valuable extensions.

In this research, we present a model that, due to computational tractability, was previously thought to be infeasible. In particular, we introduce here a generalized model for count data based upon an assumed Weibull interarrival process that nests the Poisson and negative binomial models as special cases. The computational intractability is overcome by deriving the Weibull count model using a polynomial expansion which then allows for closed-form inference (integration term-by-term) when incorporating heterogeneity due to the conjugacy of the expansion and a commonly employed gamma distribution.

In addition, we demonstrate that this new Weibull count model can: (a) sometimes alleviate the need for heterogeneity suggesting that what many think is overdispersion may just be model misfit due to a different and more flexible timing model (Weibull versus exponential), (b) model both over and under dispersed count data, (c) allow covariates to be introduced straightforwardly through the hazard function, and (d) be computed in standard software. In fact, we demonstrate the efficacy of our approach using a data analysis run, including bootstrap standard errors computed via a weighted-likelihood, run in Microsoft Excel.

Keywords: Count models, Duration models, Weibull, Overdispersion, Underdispersion

Suggested Citation

Bradlow, Eric and Fader, Peter and Adrian, Moshe and McShane, Blakeley B., Count Models Based on Weibull Interarrival Times (January 2006). Available at SSRN: https://ssrn.com/abstract=729886 or http://dx.doi.org/10.2139/ssrn.729886

Eric Bradlow (Contact Author)

University of Pennsylvania - Marketing Department ( email )

700 Jon M. Huntsman Hall
3730 Walnut Street
Philadelphia, PA 19104-6340
United States
215-898-8255 (Phone)

Peter Fader

University of Pennsylvania - Marketing Department ( email )

700 Jon M. Huntsman Hall
3730 Walnut Street
Philadelphia, PA 19104-6340
United States

Moshe Adrian

University of Illinois at Urbana-Champaign ( email )

601 E John St
Champaign, IL 61820
United States

Blakeley B. McShane

Northwestern University - Kellogg School of Management ( email )

2001 Sheridan Road
Evanston, IL 60208
United States

Register to save articles to
your library

Register

Paper statistics

Downloads
375
Abstract Views
2,353
rank
79,041
PlumX Metrics