A finite mixture latent trajectory model for hirings and separations in the labor market

Bacci, Silvia and Bartolucci, Francesco and Pigini, Claudia and Signorelli, Marcello (2014): A finite mixture latent trajectory model for hirings and separations in the labor market.

Abstract

We propose a finite mixture latent trajectory model to study the behavior of firms in terms of open-ended employment contracts that are activated and terminated during a certain period. The model is based on the assumption that the population of firms is composed by unobservable clusters (or latent classes) with a homogeneous time trend in the number of hirings and separations. Our proposal also accounts for the presence of informative drop-out due to the exit of a firm from the market. Parameter estimation is based on the maximum likelihood method, which is efficiently performed through an EM algorithm. The model is applied to data coming from the Compulsory Communication dataset of the local labor office of the province of Perugia (Italy) for the period 2009-2012. The application reveals the presence of six latent classes of firms.

• Item Type: MPRA Paper
• Language: English
• Keywords: Finite mixture models, Latent trajectory model, Compulsory communications, hirings and separations
• Subjects: C - Mathematical and Quantitative Methods > C3 - Multiple or Simultaneous Equation Models ; Multiple Variables > C33 - Panel Data Models ; Spatio-temporal Models; C - Mathematical and Quantitative Methods > C4 - Econometric and Statistical Methods: Special Topics > C49 - Other; J - Labor and Demographic Economics > J6 - Mobility, Unemployment, Vacancies, and Immigrant Workers > J63 - Turnover ; Vacancies ; Layoffs

References

1. Akaike, H.: Information theory and an extension of the maximum likelihood principle. In: Petrov, B. N., Caski, F. (eds.) Proceeding of the Second International Symposium on Information Theory, pp. 267-281. Akademiai Kiado, Budapest (1973)

2. Bollen, K.A., Curran, P.J.: Latent curve models: A structural equation perspective. Wiley, Hoboken, NJ (2006)

3. Dempster, A. P., Laird, N. M., Rubin, D. B.: Maximum likelihood from incomplete data via the EM algorithm (with discussion). Journal of the Royal Statistical Society, Series B, 39, 1–38 (1977)

4. Hijzen, A., Mondauto, L., Scarpetta, S.: The Perverse Effects of Job-Security Provisions on Job Security in Italy: Results from a Regression Discontinuity Design. IZA Discussion Paper number 7594 (2013) Available via http://ftp.iza.org/dp7594.pdf.

5. Keribin, C.: Consistent estimation of the order of mixture models. Sankhya: The Indian Journal of Statistics, Series A 62, 49–66 (2000)

6. McLachlan, G., Peel, D.: Finite mixture models. Wiley, Hoboken, NJ (2000)

7. Muth ́ n, B.: Latent variable analysis: Growth mixture modelling and related techniques for longitudinal data. In: Kaplan, D. (eds.) Handbook of Quantitative methodology for the social sciences, pp. 345-368. Sage, Newbury Park, CA (2004)

8. Muthén, B., Shedden, K.: Finite mixture modelling with mixture outcomes using the EM algorithm. Biometrics 55, 463–469 (1999)

9. Schivardi, F., Torrini, R.: Identifying the effects of firing restrictions through size-contingent differences in regulation. Labour Economics 15, 482–511 (2008)

10. Schwarz, G.: Estimating the dimension of a model. The Annals of Statistics 6, 482–511 (1978)