Indirect Inference (I-I) is a popular technique for estimating complex parametric models whose likelihood function intractable, however, the statistical efficiency of I-I estimation questionable. While efficient method moments, Gallant and Tauchen (1996), promises efficiency, price to pay this loss parsimony thereby potential lack robustness model misspecification. This stands in contrast simpl...

We establish strong consistency and asymptotic normality of the maximum likelihood estimator for stochastic time-varying parameter models driven by score predictive conditional function. For this purpose, we formulate primitive conditions global identification, invertibility, consistency, both under correct specification misspecification model. A detailed illustration is provided a volatility m...