Forecasting Model of Monthly Gold Ornament Selling Prices

Abstract

Abstract

The objective of this research is to construct the appropriate forecasting model of the gold ornament prices by the methods of Box-Jenkins, Holt???s exponential smoothing, and damped trend exponential smoothing. Time series data consist of the average selling prices per month of gold ornament obtained from the website of Gold Trades Association during January 2007 to December 2012 with total 72 values. The data are divided into two series; the first is from January 2007 through May 2012 with 65 values used to construct the models, the second is from June through December 2012 with 7 values used to examine the accuracy of forecasting models, using mean absolute percentage error (MAPE) as a criterion of comparison. The results show that the Holt???s exponential smoothing method is the most suitable for this time series and the forecasting model is ??where m represents the number of periods to forecast ahead with the starting value is June 2012 (m = 1).

??

Keywords: Gold Ornament, Box-Jenkins, Holt???s Exponential Smoothing, Damped Trend

Exponential Smoothing, Mean Absolute Percentage Error