The Pennsylvania reemployment bonus experiments : how a survival model helps in the analysis of the data

Schunk, Daniel

dp03_35.pdf - Published

Download (263kB)

URN: urn:nbn:de:bsz:180-madoc-27486
Document Type: Working paper
Year of publication: 2003
The title of a journal, publication series: None
Publication language: English
Institution: School of Law and Economics > Sonstige - Fakultät für Rechtswissenschaft und Volkswirtschaftslehre
MADOC publication series: Sonderforschungsbereich 504 > Rationalitätskonzepte, Entscheidungsverhalten und ökonomische Modellierung (Laufzeit 1997 - 2008)
Subject: 330 Economics
Subject headings (SWD): USA , Pennsylvania , Stellensuche , Arbeitslosenversicherung , Versicherungsprämie , Wirkungsanalyse , Mikroökonomie , Schätzung
Keywords (English): Reemployment experiments , survival analysis , quantile regression
Abstract: Survival models for life-time data and other time-to-event data are widely used in many fields, including medicine, the environmental sciences, engineering etc. They have also found recognition in the analysis of economic duration data. This paper provides a reanalysis of the Pennsylvania Reemployment Bonus Experiments, which were conducted in 1988-89 to examine the effect of different types of reemployment bonus offers on the unemployment spell. A Cox-proportional-hazards survival-model is fitted to the data and the results are compared to the results of a linear regression approach and to the results of a quantile regression approach. The Cox-proportional-hazards model provides for a remarkable goodness of fit and yields less effective treatment responses, therefore lower expectations concerning the overall implications of the Pennsylvania experiment. An influence analysis is proposed for obtaining qualitative information on the influence of the covariates at different quantiles. The results of the quantile regression and of the influence analysis show that both the linear regression and the Cox-model still impose stringent restrictions on the way covariates influence the duration distribution, however, due to its flexibility, the Cox-proportional hazards model is more appropriate for analysing the data.
Additional information:

Das Dokument wird vom Publikationsserver der Universitätsbibliothek Mannheim bereitgestellt.

Metadata export


+ Search Authors in

+ Download Statistics

Downloads per month over past year

View more statistics

You have found an error? Please let us know about your desired correction here: E-Mail

Actions (login required)

Show item Show item