Abstract
The high volatility of the price of natural rubber (NR) posts a significant risk to producers, traders, consumers, and others involved in the production of NR. Thus, it is crucial for decision makers to statistically and accurately project the price of NR. This research uses univariate and multivariate econometric models to forecast the short-run average monthly price of Standard Malaysia Rubber 20 (SMR20), using monthly data from January 2000 to September 2011. The autoregressive integrated moving average (ARIMA) and vector autoregressive (VAR) or vector error correction (VEC) models will be employed in the analysis for forecasting. The study also generates an out-of-sample forecast to analyze and compare the statistical results from all the models in order to determine the accuracy of which methods are more accurate in terms of statistical criteria and visual proximity with the actual prices. The results show that multivariate time series models outperform univariate time series models in term of forecasting accuracy.