Abstract:
The objective of this research is to study and compare five methods of forecasting time series, namely WMA, H-WEMA, H-WEMA Winters, H-WEMA Winters with ARMA model and ARIMA model. In this study, we use the monthly time series data of natural gas consumption in Thailand. The forecasting time series were separated into two duration times. The first is the forecasting time series data in previous time form January 1997 to December 2016 for 240 months. This data was used for selection the suitable models in each forecasting methods. As the second is the lead forecasting data form January 2017 to October 2017 for 10 months, which used for selection the most suitable model. The Mean Absolute Percentage Error (MAPE) and R-Squared are used as the comparative criteria. It was found that the H-WEMA Winters with ARMA model is the most suitable model for forecasting the monthly time series data of natural gas consumption in Thailand. The H-WEMA Winters with ARMA model for 10 months lead forecast had the lowest MAPE of 2.6859 percent with the highest R-Squared of 99.1270 percent .