An Alternative Estimation Method of a Time-Varying Parameter Model

TL;DR

This paper introduces a non-Bayesian, regression-based estimation method for time-varying AR models that avoids Kalman filtering, reduces the pile-up problem, and handles complex features like stochastic volatility and structural breaks.

Contribution

It proposes a novel GLS-based approach that simplifies estimation of time-varying parameters and extends applicability to non-Gaussian errors and models with structural changes.

Findings

Efficient estimation without Kalman filtering

Negligible pile-up problem compared to maximum likelihood

Applicable to models with stochastic volatility and structural breaks

Abstract

A non-Bayesian, regression-based or generalized least squares (GLS)-based approach is formally proposed to estimate a class of time-varying AR parameter models. This approach has partly been used by Ito et al. (2014, 2016a,b), and is proven to be efficient because, unlike conventional methods, it does not require Kalman filtering and smoothing procedures, but yields a smoothed estimate that is identical to the Kalman-smoothed estimate. Unlike the maximum likelihood estimator, the possibility of the pile-up problem is negligible. In addition, this approach enables us to deal with stochastic volatility models, models with a time-dependent variance-covariance matrix, and models with non-Gaussian errors that allow us to deal with abrupt changes or structural breaks in time-varying parameters.