Forecasting Model for the Retail Prices of Gasohol 95 in Bangkok and Peripheral Areas

Abstract

The objective of this study was to construct an appropriate forecasting model for the daily retail prices of gasohol 95 in Bangkok and peripheral areas. The data used in this study is the secondary data from the website of PTT Public Company Limited (Thailand) during January 1, 2012 to February 25, 2013 (422 values). The time-series data were divided into two categories. The first category has 412 values, which are price data from January 1, 2012 until February 15, 2013 for the modeling by the methods of Box-Jenkins, Holt’s exponential smoothing, Brown’s exponential smoothing, and damped trend exponential smoothing. The second category has 10 values, which are price data from February 16 to 25, 2012 for checking the accuracy of the forecasting models via the criterion of the lowest mean absolute percentage error (MAPE). The results showed that Brown’s exponential smoothing is the most suitable for this time series and the forecasting model is: