Microeconometric Modelling of Unemployment Durations under Consideration of Macroeconomic Variations

Microeconometric Modelling of Unemployment Durations under Consideration of Macroeconomic Variations

Since many years the analysis of the determinants for the length of individual unemployment durations is subject to scientific and political discussion in Germany. However, the results of the existing scientific contributions are controversial and some of them did not treat important aspects. In this project new developments in econometrics for duration analysis are considered and extended for an application to Germany. As data we use a sample of the German social security records (IAB-Beschäftigtenstichprobe) which was rarely used in this field in the past. It is our goal to use a non- or semiparametric specification that accounts for macroeconomic variations and the distribution of unobserved heterogeneity in order to obtain a flexible model with low risk of misspecification. We considerMixed-Proportional-Hazard models with recent extensions andquantile regression estimators as our main methods. The analysis should result in a deep empirical knowledge of the micro- and macroeconomic determinants of unemployment durations in Germany. Such assured facts may result in recommendations for the labor market policy in Germany.

Project members

Bernd Fitzenberger

Bernd Fitzenberger

Project Coordinator
Research Associate

To the profile
Gerard J. van den Berg

Gerard J. van den Berg

Research Associate

To the profile
Client/Allowance
Cooperation partner
Institute for Employment Research (IAB), Nürnberg, DE // University College London, London, UK // Tinbergen Institute, Rotterdam, NL // Universidade Nova de Lisboa, Lissabon, PT // Government Institute for Economic Research, Helsinki, FI

Selected Publications

Using Quantile Regression for Duration Analysis

Fitzenberger, Bernd and (2005), Using Quantile Regression for Duration Analysis, ZEW Discussion Paper No. 05-65, Mannheim.

A Note on Implementing Box-Cox Quantile Regression

Fitzenberger, Bernd, and Xuan Zhang (2004), A Note on Implementing Box-Cox Quantile Regression, ZEW Discussion Paper No. 04-61, Mannheim.

Contact