Machine learning for labour market matching

Arbeitspapier

This paper develops a large-scale application to improve the labour market matching process with model- and algorithm-based statistical methods. We use comprehensive administrative data on employment biographies covering individual and job-related information of workers in Germany. We estimate the probability that a job seeker gets employed in a certain occupational field. For this purpose, we make predictions with common statistical methods (OLS, Logit) and machine learning (ML) methods. The findings suggest that ML performs best regarding the out-of-sample classification error. In terms of the unemployment rate hypothetically, the advantage of ML compared to the common statistical methods would stand for a diference of 0.3 - 1.0 percentage points.

Object Details

• Language: Englisch

References and Relationships

• Bibliographic citation: Series: IAB-Discussion Paper ; No. 3/2022

Classification and Topics

• Classification: Wirtschaft ; Semiparametric and Nonparametric Methods: General ; Neural Networks and Related Topics ; Large Data Sets: Modeling and Analysis ; Unemployment: Models, Duration, Incidence, and Job Search
• Subject: Labour Market ; Machine Learning ; Matching ; Random Forest

Contributors, Places and Time

• Event: Geistige Schöpfung
• (who): Mühlbauer, Sabrina ; Weber, Enzo
• Event: Veröffentlichung
• (who): Institut für Arbeitsmarkt- und Berufsforschung (IAB)
• (where): Nürnberg
• (when): 2022

Further information

• DOI: doi:10.48720/IAB.DP.2203
• Handle: http://hdl.handle.net/10419/253713
• Last update: 10.03.2025, 11:41 AM CET