Yip, Chee Yin and Lim, Hock Eam (2012) An empirical investigation on the forecasting ability of Mallows Model Averaging in a macro economic environment. In: International Conference on Fundamental and Applied Sciences 2012, 12th to 14th June 2012, Kuala Lumpur, Malaysia.

PDF Restricted to Repository staff only Download (136kB) | Request a copy

Abstract

This paper investigates the forecasting ability of Mallows Model Averaging (MMA) by conducting an empirical analysis of five Asia countries, Malaysia, Thailand, Philippines, Indonesia and China's GDP growth rate.Results reveal that MMA has no noticeable differences in predictive ability compared to the general autoregressive fractional integrated moving average model(ARFIMA) and its predictive ability is sensitive to the effect of financial crisis. MMA could be an alternative forecasting method for samples without recent outliers such as financial crisis.

Item Type:	Conference or Workshop Item (Paper)
Additional Information:	AIP Conference Proceedings ISSN: 0094-243X Organized by: Universiti Teknologi PETRONAS
Uncontrolled Keywords:	Mallows Model Averaging; Predictive ability; Model selection; Model Averaging.
Subjects:	H Social Sciences > HB Economic Theory Q Science > QA Mathematics > QA76 Computer software
Divisions:	College of Arts and Sciences
Depositing User:	Dr. Chee Yin Yip
Date Deposited:	23 Oct 2014 07:19
Last Modified:	23 Oct 2014 07:19
URI:	https://repo.uum.edu.my/id/eprint/12415

Actions (login required)

View Item