Statistical Modeling of Duration Dependence in Business Cycles Using a Modulated Power Law Process

42 Pages Posted: 4 May 2005

See all articles by Haigang Zhou

Haigang Zhou

Cleveland State University - Nance College of Business Administration

Steven E. Rigdon

Southern Illinois University at Edwardsville - Department of Mathematics & Statistics

Multiple version iconThere are 2 versions of this paper

Date Written: November 2004

Abstract

Applying the modulated power law process, a generalized model which is a compromise between a renewal process and a nonhomogeneous Poisson process, we find the presence of positive duration dependence in all samples, although the magnitudes vary from sample to sample. A goodness-of-fit test rejects the renewal process assumption which was used in many previous studies, and suggests that a more general model such as the modulated power law process is needed.

Keywords: Weibull process, renewal process, maximum likelihood estimation, buinsess cycle

Suggested Citation

Zhou, Haigang and Rigdon, Steven E., Statistical Modeling of Duration Dependence in Business Cycles Using a Modulated Power Law Process (November 2004). Available at SSRN: https://ssrn.com/abstract=714061 or http://dx.doi.org/10.2139/ssrn.714061

Haigang Zhou (Contact Author)

Cleveland State University - Nance College of Business Administration ( email )

2121 Euclid Avenue
Department of Finance, BU 215
Cleveland, OH 44115-2214
United States

Steven E. Rigdon

Southern Illinois University at Edwardsville - Department of Mathematics & Statistics ( email )

1 Hairpin Drive
Edwardsville, IL 62026-1102
United States