A New Approach to the Economic Analysis of Nonstationary Time Series and the Business Cycle

TLDR

Nonstationary time series often experience discrete regime shifts that are unobserved, requiring probabilistic inference about their timing and magnitude. The study proposes a tractable method for modeling regime changes in nonstationary series. An algorithm based on a nonlinear iterative filter is introduced to perform the probabilistic inference.

Abstract

This paper proposes a very tractable approach to modeling changes in regime. The parameters of an autoregression are viewed as the outcome of a discrete-state Markov process. For example, the mean growth rate of a nonstationary series may be subject to occasional, discrete shifts. The econometrician is presumed not to observe these shifts directly, but instead must draw probabilistic inference about whether and when they may have occurred based on the observed behavior of the series. The paper presents an algorithm for drawing such probabilistic inference in the form of a nonlinear iterative filter