B1264
Title: Time series forecasting using hybrid regime switching ANN models
Authors: Jie Cheng - Keele University (United Kingdom) [presenting]
Abstract: Regime-switching models have been widely used for both economic and financial time series. However, their out-of-sample forecasting performance is frequently inferior to simple benchmark models for standard loss functions. The purpose is to determine whether improvements can be achieved in forecasting performance by using a hybrid regime-switching model with a set of Machine Learning techniques and incorporating financial and/or economic variables. The empirical results with some real data sets indicate the effectiveness of the new combinatorial model in obtaining more accurate forecasting as compared to existing regime-switching models.
