Do the Markov Switching-based Hybrid Models Perform Better in Forecasting Exchange Rates?

Abstract

ABSTRACT In this study, we extend the traditional monetary model and the random walk model with Markov-switching method and propose two new forecasting models called the Markov switching monetary model (MSMM) and Markov switching random walk model (MSRW). Then, we evaluate the forecasting ability of these two new mixed models, MSMM and MSRW, and compare their performance with the traditional pure monetary model and pure Random walk model based on Mean Squared Forecast Error and Mean Absolute Forecast Error. The results show that the two hybrid models significantly improve the forecasting ability compared with the two traditional models in most scenarios. Moreover, we reexamine the role of data frequency in determining the number of regimes and in affecting the accuracy of forecast evaluation with different data frequency.