Title: Time-varying GMM estimation in structural time series models
Authors: Yu Bai - Bocconi University (Italy) [presenting]
Abstract: The aim is to develop time-varying continuously updated GMM estimation and inferential theory for moment conditional models whose coefficients vary stochastically and smoothly over time. We propose two new tests of structural stability in this context. After deriving the asymptotic properties of the estimators and test statistics, we assess finite sample performance by an extensive Monte-Carlo study and illustrate their application by an empirical example on conditional asset pricing models with stochastic discount factor representation.