Econometric Forecasting Models for Short Term Natural Rubber Prices

This study presents a number of short-term ex-post forecasts of single equation model, Multivariate Autoregressive Moving Average (MARMA) model, simultaneous supply-demand and price system equation model, and Autoregressive Integrated Moving Average (ARIMA) model, and ARCH-type models of natural rubber (NR) SMR20 (Standard Malaysia Rubber of grade 20) prices in the world NR market. The ARCH-type models (Autoregressive Conditional Heteroskedasticity) used include the GARCH (1,1) (Generalized ARCH) model, EGARCH (1,1) (The Exponential GARCH) model, PARCH (1,1) (The Power ARCH) and CGARCH (1,1) (The Component GARCH) model. These were developed for ex-post forecast of short-term monthly SMR20 prices in the world NR market. Natural rubber is a vital commodity used in the manufacture of a wide range of rubberbased products. Over 20 million families are dependent on rubber cultivation for their livelihood in the world NR market. The years 1997 to 1999 and as well as in the year 2000 were turbulent years for the economies in South-East and East Asia. In 2008, the extremely low prices due to the outbreak of the global recession. It experienced during these years contributed to price volatility and instability in many countries, especially rubber smallholders in South East Asia. Moreover, the crude petroleum oil price is an important component of synthetic rubber. A fall in the crude petroleum oil price relates to synthetic rubber. It influences a declining share of synthetic rubber in total rubber consumption, and also a weak currency exchange affects in the NR producing countries because most commodities are traded in US dollar. This could be a good reason for taking the current NR price forecasting study. It would be also a direction of short term NR price movement for policy formulation. Furthermore, the conceptual economic framework of this study was a good starting point for discussion and perceptive of shortterm ex-post forecast of NR price forecasting models developed, with the opportunity of using some of these factors later in the other study for the forecasting of rubber prices. The model specifications were developed in order to discover the inter-relationships between NR production, consumption and prices of SMR20, to forecast the NR price of SMR20 using single equation model, MARMA model, simultaneous system equation of supply-demand and price forecasting model, ARIMA model, and ARCH-type models, to analyze and compare the various NR price forecasting models individually in terms of their comparative price forecasting accuracy and to determine which between the models are more efficient. The models were utilized using monthly data from January 1990 to December 2008 as estimation period, providing a total of 228 observations and data was used as an ex-post forecasts. All data (variables) were tested for unit root test using the Augmented Dickey-Fuller (ADF) test and the Phillips-Perron (PP) test and were found to be stationary at first difference. The Granger causality test was tested for the direction of a Granger causality relationship between two variables