What is it about?

In this study, we evaluate the effect of news on consumer unemployment expectations for sixteen socio-demographic groups. By means of genetic programming we estimate symbolic regressions that link monthly unemployment rates in the Euro Area to qualitative expectations about a wide range of economic variables. We then use the evolved expressions to compute unemployment expectations for each consumer group. We first assess the out-of-sample forecast accuracy of the evolved indicators. Next, we link news about inflation, industrial production, and stock markets to unemployment expectations. With this aim we match positive and negative news with consumers’ unemployment sentiment using a distributed lag regression model for each news item.

Featured Image

Why is it important?

This is the first study to generate survey-based unemployment sentiment indicators through genetic programming. Regarding the forecasting out-of-sample performance of the evolved indicators, we obtain similar results across the different socio-demographic groups. The best forecast results are obtained for respondents between 30 and 49 years. The group where we observe the bigger differences among categories is the occupation, where the lowest forecast errors are obtained for the unemployed respondents. We also find asymmetries in the response of consumers’ unemployment expectations to economic news. Their reaction tends to be stronger in the case of negative news, especially in the case of inflation.


Empirical modelling via genetic programming is a very promising field of research. This approach is especially suitable for finding patterns in large data sets with little or no prior information about the system, and allows to simultaneously evolve both the structure and the parameters of the models without imposing any assumptions.

Oscar Claveria
AQR-IREA, Univeristy of Barcelona

Read the Original

This page is a summary of: Unemployment expectations: A socio-demographic analysis of the effect of news, Labour Economics, October 2019, Elsevier, DOI: 10.1016/j.labeco.2019.06.002.
You can read the full text:




The following have contributed to this page