Forecasting crude palm oil price movement from the dynamic relationship between CPO spot and futures prices: empirical evidence

Our study tests a forecasting model of daily CPO price movements, which exploits the information that is provided by the CPO futures market through the dynamic relationship between CPO spot and futures prices. The cost of carry model is employed to justify the existence of a long run relationship between CPO spot and futures prices. Daily data from 1996 to 2011 is used to show evidence of cointegration between CPO spot price and the price of the 1,2,3 and 4-month futures contracts. From the existence of the long run relationship, a vector error correction model (VECM) is then estimated. From the in-sample performance of VECM, we found that the information provided by the CPO futures can explain a large portion of the CPO price movements. To evaluate forecasting performance, the VECM is then compared with the random walk model (RWM) used as a benchmark and the results show that the VECM is a superior model compared to RWM in forecasting 1-month CPO futures contract price movements both in terms of point forecasting performance and also in terms of market timing ability. Further evidence was established in comparing the performance of VECM with that of a vector autoregressive model in differences (VARD). The result shows that the VECM outperforms the VARD both in-sample and out-of sample.