A0309
Title: Time-varying forecasting regressions augmented by a dynamic nonstationary factor structure
Authors: Tingting Cheng - Nankai University (China) [presenting]
Abstract: The purpose is to introduce a time-varying predictive regression augmented by nonstationary common factors, considering both static and dynamic factor models. An easy-to-implement sieve-based estimation method is proposed to estimate the unknown time-varying coefficients in the predictive regression. Large sample theories are established for those estimators and also examine their finite-sample performance by simulation studies. The advantages of this new model are further demonstrated through an empirical application to U.S. inflation forecasting.
