A1564
Title: Semi-parametric estimation of non-stationary autoregressive models
Authors: Qiyuan Wang - Texas A&M University (United States)
Giovanni Motta - Columbia University (United States) [presenting]
Abstract: A novel semi-parametric approach is developed for accurately estimating time-varying mean and variance in autoregressive (AR) models. By combining B-splines with (i) generalized least squares (GLS) estimation to account for serial correlation, and (ii) weighted least squares (WLS) for smooth parametrization, the approach addresses the challenges posed by time-varying dynamics in time series Data. The covariance matrix in the GLS estimation of the spline coefficients is iteratively updated through the WLS estimation of the AR coefficients in a band-limited manner. A new autoregressive model is proposed that incorporates time-varying variance with a finite bounded envelope function, and a novel method is introduced to estimate it through splines. Additionally, the order of the AR model is determined through a generalized Bayesian information criterion (GBICp) that incorporates prior information. The effectiveness of the methodology is demonstrated through extensive simulations and applications to Federal Reserve Economic Data. Asymptotic theory is derived for the approach and rates of convergence are compared with those achieved by existing methods.
