Non-Parametric Approach to Dynamic Time Series Discrete Choice Models

Park, Byeong U.;Simar, LĂ©opold;Zelenyuk, Valentin
(2013) , 25 pages

Files

ISBADP2013_52.pdf
  • Open Access
  • Adobe PDF
  • 494.63 KB
Download

Details

Authors
  • Park, Byeong U.Seoul National University Korea
    Author
  • Simar, LĂ©opoldUCLouvain
    Author
  • Zelenyuk, ValentinUniversity of Queensland
    Author
Abstract
Dynamic discrete choice models are very important in applied research, frequently used in parametric contexts, and here we develop a non-parametric approach to such models. The main contribution of our work is generalization of the non-parametric quasi-likelihood method to the context that allows for time series models, and in particular with lags of the (discrete) dependent variable appearing among regressors. We show consistency and asymptotic normality of the estimator for such models and illustrate it with simulated and real data examples.
Affiliations

Citations

  • APA
  • Chicago
  • FWB

Park, B. U., Simar, L., & Zelenyuk, V. (2013). Non-Parametric Approach to Dynamic Time Series Discrete Choice Models (ISBA Discussion Paper 2013/52). https://hdl.handle.net/2078.5/199625