A0239
Title: Estimating latent-variable panel data models using parameter-expanded SEM methods
Authors: Siqi Wei - IE University (Spain) [presenting]
Abstract: New estimation algorithms are presented for three types of dynamic panel data models with latent variables: factor models, discrete choice models, and persistent-transitory quantile processes. The new methods combine the parameter expansion (PX) ideas in a prior study with the stochastic expectation-maximization (SEM) algorithm in likelihood and moment-based contexts. The goal is to facilitate convergence in models with a large space of latent variables by improving algorithmic efficiency. This is achieved by specifying expanded models within the M step. Effectively, new estimators for the pseudo-data are proposed within iterations that take into account the fact that the model of interest is misspecified for draws based on parameter values far from the truth. Conditions are provided under which the new algorithm dominates SEM in terms of the global rate of convergence and characterizes the asymptotic distributions of the estimators based on PX-SEM algorithms. Finally, in simulations, it is shown that the new algorithms significantly improve the convergence speed relative to standard SEM algorithms, sometimes dramatically so.
