Author(s): Quang Hung Do, Tran Thi Hoang Yen,
The fluctuations in primary commodity prices have a significant impact on global economy. Therefore, forecasting price of major commodities prices has been getting much attention both from academic and practitioners' communities. The objective of this study is to develop a model based on group method of data handling (GMDH) technique in one day-ahead forecasting the market prices for major commodities including copper, crude oil, gas and silver. The data on commodities trading were collected from January 2000 to October 2019. In order to validate the effectiveness of the proposed model, other models based on adaptive neuro fuzzy inference system (ANFIS), artificial neural network (ANN), long short-term memory (LSTM) were also developed. The performance indexes including RMSE, MAPE, MAE, R and Theil’s U were used to make comparison of the models. The results showed that the proposed model based on GMDH technique outperforms than other methods in prediction of commodity prices. The GMDH-based model provides a promising alternative for price prediction. The GMDH can be a useful tool for economists and practitioners dealing with the forecasting of the commodity price.