A Comparison of Forecasting Methods between Box-Jenkins and Holt’s Exponential Smoothing Methods for Predicting the Retail Prices of Hip Meat

Abstract

The purpose of this research is to compare two forecasting methods which are BoxJenkins method and Holt’s exponential smoothing method for predicting the retail prices of hip meat. Time series that used is the monthly data from the website of Department of Internal Trade of Thailand during January 2004 to June 2013 (114 values). The data are split into two sets, the first 102 values from January 2004 until June 2012 for the modeling and the last 12 values from July 2012 until June 2013 for finding the most suitable forecasting method by the criteria of the lowest mean absolute percentage error and mean squared error. The results show
that Holt’s exponential smoothing method is suitable for this time series than Box-Jenkins method. The forecasting model that constructed from the total time series data 114 values is _t+m = 132.49997 + 0.39578 (m)  where m represents the number of months to forecast ahead with the predictive value start at July 2013 (m = 1).