The paper presents various econometric models to forecast the short-run monthly prices
of SMR20 in the world market. These models of the world natural rubber have
specifications for production, consumption, prices. Among the objectives of this study are
to determine the inter-relationships between NR production, consumption and prices, to
derive and forecast the multivariate autoregressive-moving average (MARMA) model which
integrates the normal autoregressive-integrated-moving average (ARIMA) and econometric
model and to estimate and test individually in terms of their comparative price
forecasting accuracy. The study develops some short term ex ante and ex post forecasts
of natural rubber (NR) prices in the world market. The models are then used to generate
ex ante forecasts for the period of January 2007 to December 2010. The results show that
MARMA’s ex post forecasts are more efficient in terms of its statistical criteria or even
by visual proximity with the actual prices. The forecast figures reveals important
implications for model specification and forecasting in relation to the current and expected
fundamentals of the natural rubber producers and consumers as well as traders and
planners for new investment decisions in the natural rubber world market.