Understanding the volatility model of ARCH and GARCH using Standard Malaysian Rubber 20 (SMR 20) prices

Abstract

Natural rubber is one of Malaysia's most important agricultural commodities and contributes significantly to the country's export earnings and economic development. However, fluctuations in rubber prices create uncertainty for producers, traders, investors and policymakers, making volatility analysis an essential area of study. This paper examines the volatility characteristics of Standard Malaysian Rubber Grade 20 (SMR20) prices using the Autoregressive Conditional Heteroscedasticity (ARCH) model and its generalised extension (GARCH). Daily price data covering January 2023 to December 2025 were obtained from the Malaysian Rubber Council (MRC), yielding 726 observations after data cleaning. The price series was transformed into logarithmic returns to satisfy the stationarity requirement prior to model estimation. The ARCH-LM test confirmed the presence of significant ARCH effects in the return series, establishing volatility clustering and conditional heteroskedasticity. Six specifications, namely ARCH(1), ARCH(2), ARCH(3), GARCH(1,1), GARCH(1,2) and GARCH(1,3), were estimated and compared using the Akaike Information Criterion (AIC) and the Schwarz Information Criterion (SIC). The results show that the GARCH(1,1) model records the lowest AIC (3.136771) and SIC (3.162046) values among all six specifications, making it the most appropriate model for capturing the conditional volatility of SMR20 prices. Two major volatility episodes were identified, in the third quarter of 2024 and the first half of 2025, corresponding to the sharp price rally and the subsequent correction. These findings provide useful information for rubber industry stakeholders in relation to price risk management and forecasting.