How well can Noncognitive Skills Predict Unemployment? A Machine Learning Approach
Type
working paper
Date Issued
2019-04
Author(s)
Abstract
We study the predictive quality of noncognitive skill measures by means of machine learning techniques. Unlike previous empirical approaches centering around the in-sample explanatory power of noncognitive skills, our approach focuses on the performance of predicting individual unemployment over a long-run horizon of 20 years and more. Our machine learning approach can cope not only with the challenge of selecting the most relevant factors from data with
a large number of skill measures but also leads to a sparse set of skill measures which is economically and psychologically interpretable. Using data from the British Cohort Study (BCS), we compare the predictive power of different noncognitive skill measures and illustrate, how our estimates can be used to optimize the assignment mechanisms for manpower training programs and psychological intervention schemes for youths and young adults.
a large number of skill measures but also leads to a sparse set of skill measures which is economically and psychologically interpretable. Using data from the British Cohort Study (BCS), we compare the predictive power of different noncognitive skill measures and illustrate, how our estimates can be used to optimize the assignment mechanisms for manpower training programs and psychological intervention schemes for youths and young adults.
Keywords
noncognitive skills
unemployment
machine learning
group lasso
File(s)![Thumbnail Image]()
Name
02-Mareckova-Pohlmeier-Noncognitive-Skills-predicting-Unemployment.pdf
Size
1.21 MB
Format
Adobe PDF
Checksum (MD5)
e430cd87d631344819970b22112cfc37