Monte Carlo evidence on the estimation of AR(1) panel data sample selection models

• Sergi Jiménez Martín ^[2] ; José María Labeaga ^[1]
  1. [1] Universidad Nacional de Educación a Distancia

     Universidad Nacional de Educación a Distancia

     Madrid, España

     
  2. [2] Universitat Pompeu Fabra

     Universitat Pompeu Fabra

     Barcelona, España

     
• Localización: Documentos de trabajo ( FEDEA ), ISSN 1696-7496, Nº. 1, 2016, págs. 1-17
• Idioma: inglés
• Enlaces
  • Texto completo (pdf)
• Resumen

  • We present Generalized Method of Moments estimators for AR(1) dynamic panel data sample selection models. We perform a Monte Carlo study to evaluate the finite sample properties of the proposed estimators. Our results suggest that correcting for sample selection in many standard cases does not add much to the uncorrected estimates. In particular, the magnitude of the biases is similar and very small when estimating the model either correcting or not the equation of interest. This equivalence also holds in the dynamic model with exogenous regressors. These results are especially relevant for practitioners either when there is selection of unknown form or selection is difficult to model.