# Predicting Unemployment with Machine Learning Based on Registry Data

- 33 Downloads

## Abstract

Many statistical models have been developed to understand the causes of unemployment, but predicting unemployment has received less attention. In this study, we develop a model to predict the labour market state of a person based on machine learning trained with a large administrative unemployment registry. The model specifies individuals as Markov chains with person specific transition rates. We evaluate the model on three tasks, where the goal is to predict who has the highest risk of escaping unemployment, becoming unemployed, and being unemployed at any given time. We obtain good performance (AUC: 0.80) for the machine learning model of lifetime unemployment, and very good performance (AUC: 0.90+) to the near future when we know the recent labour market state of a person. We find that person information affects the predictions in an intuitive way, but there still are significant differences that can be learned by utilizing labour market histories.

## Keywords

Unemployment Machine learning Prediction## References

- 1.Ernst, E., Rani, U.: Understanding unemployment flows. Oxford Rev. Econ. Pol.
**27**(2), 268–294 (2011)CrossRefGoogle Scholar - 2.Shimer, R.: Reassessing the ins and outs of unemployment. Rev. Econ. Dyn.
**15**(2), 127–148 (2012)CrossRefGoogle Scholar - 3.Ahn, H.J., Hamilton, J.D.: Heterogeneity and unemployment dynamics. J. Bus. Econ. Stat. 1–26 (2019)Google Scholar
- 4.Honkanen, P.: Odotelaskelmat työllisyyden, työttömyyden ja eläkeajan arvioinnissa. KELA Working Papers, No. 137 (2018)Google Scholar
- 5.Pedersen, P.J., Westergård-Nielsen, N.C.: Unemployment. A review of the evidence from panel data. In: Economics of Unemployment. Edward Elgar Publishing (2000)Google Scholar
- 6.Wanberg, C.R.: The individual experience of unemployment. Ann. Rev. Psychol.
**63**, 369–396 (2012)CrossRefGoogle Scholar - 7.Kettunen, J.: Education and unemployment duration. Econ. Educ. Rev.
**16**(2), 163–170 (1997)CrossRefGoogle Scholar - 8.Ollikainen, V.: The determinants of unemployment duration by gender in Finland. VATT Discussion Papers, No. 316 (2003)Google Scholar
- 9.Kyyrä, T.: Partial unemployment insurance benefits and the transition rate to regular work. Eur. Econ. Rev.
**54**(7), 911–930 (2010)CrossRefGoogle Scholar - 10.Rokkanen, M., Uusitalo, R.: Changes in job stability: evidence from lifetime job histories. IZA Discussion Papers, No. 4721 (2010)Google Scholar
- 11.Asplund, R.: Unemployment among finnish manufacturing workers. Who gets unemployed and from where? ETLA Discussion Papers, No. 711 (2000)Google Scholar
- 12.Eriksson, T., Pehkonen, J.: Unemployment flows in Finland, 1969–95: a time series analysis. Labour
**12**(3), 571–593 (1998)CrossRefGoogle Scholar - 13.Peltola, M.: Työmarkkinasiirtymät Suomessa. Työllisyyden päättymisen jälkeinen työmarkkinasiirtymien dynamiikka vuosina 1995–1999. VATT Discussion Papers, No. 360 (2005)Google Scholar
- 14.Heckman, J.J., Borjas, G.J.: Does unemployment cause future unemployment? Definitions, questions and answers from a continuous time model of heterogeneity and state dependence. Economica
**47**(187), 247–283 (1980)CrossRefGoogle Scholar - 15.Flinn, C.J., Heckman, J.J.: New methods for analyzing individual event histories. Sociol. Methodol.
**13**, 99–140 (1982)CrossRefGoogle Scholar - 16.Mühleisen, M., Zimmermann, K.F.: A panel analysis of job changes and unemployment. Eur. Econ. Rev.
**38**(3–4), 793–801 (1994)CrossRefGoogle Scholar - 17.D’Amuri, F., Marcucci, J.: The predictive power of Google searches in forecasting US unemployment. Int. J. Forecast.
**33**(4), 801–816 (2017)CrossRefGoogle Scholar - 18.Tuhkuri, J.: ETLAnow: a model for forecasting with big data-forecasting unemployment with Google searches in Europe. No. 54. ETLA Report (2016)Google Scholar
- 19.Katris, C.: Prediction of unemployment rates with time series and machine learning techniques. Comput. Econ.
**55**, 673–706 (2019). https://doi.org/10.1007/s10614-019-09908-9CrossRefGoogle Scholar - 20.de Troya, Í.M.R., et al.: Predicting, explaining and understanding risk of long-term unemployment. In: 32nd Conference on Neural Information Processing Systems (2018)Google Scholar
- 21.Kütük, Y., Güloğlu, B.: Prediction of transition probabilities from unemployment to employment for Turkey via machine learning and econometrics: a comparative study. J. Res. Econ.
**3**(1), 58–75 (2019)Google Scholar - 22.Beyersmann, J., Allignol, A., Schumacher, M.: Competing Risks and Multistate Models with R. Springer, Heidelberg (2011). https://doi.org/10.1007/978-1-4614-2035-4CrossRefzbMATHGoogle Scholar
- 23.Tutz, G., Schmid, M.: Modeling Discrete Time-to-Event Data. Springer, Cham (2016). https://doi.org/10.1007/978-3-319-28158-2CrossRefzbMATHGoogle Scholar
- 24.Duchateau, L., Janssen, P.: The Frailty Model. Springer, Heidelberg (2007). https://doi.org/10.1007/978-0-387-72835-3CrossRefzbMATHGoogle Scholar
- 25.Cook, R.J., Lawless, J.: The Statistical Analysis of Recurrent Events. Springer, Heidelberg (2007). https://doi.org/10.1007/978-0-387-69810-6CrossRefzbMATHGoogle Scholar
- 26.Rausand, M., Høyland, A.: System Reliability Theory: Models, Statistical Methods, and Applications, vol. 396. Wiley, Hoboken (2003)zbMATHGoogle Scholar