Forecasting the Monthly Consumer Price Index of Thailand: A Comparative Study of Box-Jenkins???s Method, Winter???s Method and the Combined Forecast Method

Abstract

The objective of this research is to forecast monthly the consumer price index of
Thailand and to compare three methods of forecasting. The methods are the Box-Jenkins???s method,
Winter???s method and the combined forecast based on the regression method. The method which
gives the lowest Mean Absolute Percent Error (MAPE) is the most suitable method. Results show
that the Winter???s method is the most suitable method.

Keywords: Consumer Price Index, Forecasting, Time Series