Predicting Unemployment: What's Important?

Author Information

Robert WinslowFollow

Start Date

October 2024

End Date

October 2024

Location

Beacom Hall Room 208

Presenter Information

Robert Martin Winslow, University of South Dakota

Submission Type

Abstract

Track

Economics

Abstract

More and more, economic research makes uses of administrative panel data, which provide long time-series of highly accurate data, but lack some of the supplemental questions found in surveys. I use data from the Current Population Survey, which includes a short panel, to test whether these supplemental survey questions intrinsically contain unique information or serve as proxies for information that can also be gleaned from panel data. In particular, I forecast an individual worker's future unemployment status with and without information on each individual's recent past, and decompose the variance of the predictions in each case.

This document is currently not available here.

Share

COinS
 
Oct 5th, 10:00 AM Oct 5th, 10:50 AM

Predicting Unemployment: What's Important?

Beacom Hall Room 208

More and more, economic research makes uses of administrative panel data, which provide long time-series of highly accurate data, but lack some of the supplemental questions found in surveys. I use data from the Current Population Survey, which includes a short panel, to test whether these supplemental survey questions intrinsically contain unique information or serve as proxies for information that can also be gleaned from panel data. In particular, I forecast an individual worker's future unemployment status with and without information on each individual's recent past, and decompose the variance of the predictions in each case.