Open AccessModeling and Forecasting of Volatility using ARMA-GARCH: Case Study on Malaysia Natural Rubber Prices
Author(s) -
Intan Martina Md Ghani,
H. Abdul Rahim
Publication year - 2019
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/548/1/012023
Subject(s) - autoregressive conditional heteroskedasticity , volatility (finance) , econometrics , natural rubber , volatility clustering , heteroscedasticity , economics , autoregressive–moving average model , statistics , autoregressive model , mathematics , chemistry , organic chemistry
Malaysia is one of the largest producers of natural rubber in the world. Among the various types of natural rubber which contribute to the country’s agricultural sector is the Standard Malaysian Rubber Grade 20 (S.M.R 20). Since 2008, the rubber price has received attention of investors and Malaysia Rubber Board due to price fluctuation. The price of rubber is characterized by the existence of heavy tails and volatility clustering. These properties play a significant impact on parameter estimation and forecasting performance resulting from S.M.R 20 rubber price data. The approach used in modeling S.M.R 20 rubber price data, is Generalized Autoregressive Conditional Heteroscedasticity (GARCH) model. The aims of this paper are to find the best ARMA-GARCH model by using different specifications structures and to forecast the daily price for 20 days ahead. There are 20 models produced from different specifications in ARMA(R,M) dan GARCH(p,q) models. In this study, 1953 daily price data of S.M.R 20 are taken into consideration. The validity comparison of diagnostic checking and forecasting performance are based on AIC, AICC, SBC, HQC, MSE, RMSE and MAPE. The results reveals that ARMA(1,0)-GARCH(1,2) model is the best volatility modeling in S.M.R 20 rubber price. Based on the implications of the results, the scope of the future research directions has been widen.