Competing Risks Analysis on Predicting Times to Commit Crimes

by Jenq-Daw Lee and Cheng K. Lee.

Abstract: In this paper, rather than exploring criminological theory, we focus on methodological aspects and apply a trivariate Weibull survival model using competing risks concept to predict recidivism of committing 3 types of crimes sex, violent and others. The assumption of independence of time to commit each type of crimes is relaxed so that the association of the time to recidivism between any two types of crimes can be evaluated. We found that the correlation of time to recidivism between sex crimes and violent crimes are more correlated than other pairs of crimes. As to an individual after release, he or she has the highest probability of experiencing a charged arrest of other crimes followed by the probability of experiencing a charged arrest of violent crimes and, then, the probability of experiencing a charged arrest of sex crimes.

Key Words: survival analysis; competing risks; trivariate Weibull

Authors:
Jenq-Daw Lee, leewien@ms4.hinet.net
Cheng K. Lee, lee.abcde@gmail.com

Editor: Weiming Ke, Weiming.Ke@sdstate.edu

READING THE ARTICLE: You can read the article in portable document (.pdf) format (150664 bytes.)

NOTE: The content of this article is the intellectual property of the authors, who retains all rights to future publication.

This page has been accessed 2085 times since SEPTEMBER 21, 2009.


Return to the InterStat Home Page.