The purpose of this research was to construct the appropriate forecasting model for the export values of alcoholic beverages using three-time series analysis methods including Box-Jenkins method damped trend exponential smoothing method and combined forecasting method. Time series data from the website of Office of Agricultural Economics with total 199 months from January 1998 to July 2014 were used and divided into two series. The first 192- month data from January 1998 until December 2013 were used to build the forecasting models and the last 7-month data from January to July 2014 were used to compare the forecasting method accuracy via the criteria of the lowest mean absolute percentage error and root mean squared error. Research findings indicated that for all forecasting methods that had been studied combined forecasting method was the most accurate method and this model was when and represent the single forecasts at time t from Box-Jenkins method and damped trend exponential smoothing method respectively.

The purpose of this research was to construct an appropriate forecasting model for the export quantity of frozen and chilled chicken using three time series analysis methods including Box-Jenkins method Holt’s exponential smoothing method and damped trend exponential smoothing method. Time series data from the website of the Office of Industrial Economics with total 130 months during January 2004 to October 2014 were used and they were divided into two series. Fristly 120 months from January 2004 to December 2013 were used to build the forecasting models and lastly 10 months from January until October 2014 were used to compare the forecasting methods accuracies via the criteria of the lowest mean absolute percentage error and root mean squared error. Research findings indicated that all forecasting methods that had been studied damped trend exponential smoothing method were the most accurate method and showed the forecasting model as where m represented the number of periods to forecast ahead with the starting value was January 2014 (m = 1).

The objective of this research is to construct the appropriate forecasting model for the export quantity of concentrated latex by the three time series analysis methods which are Box-Jenkins method Winters’ multiplicative exponential smoothing method and combined forecasting method. Time series data from the website of Office of Agricultural Economics during January 1998 to February 2014 (194 values) are used and divided into two series. The first 182 values from January 1998 until February 2013 for the models prediction and the last 12 values from March 2013 until February 2014 for comparison the performance of forecasting models via the criterion of the lowest root mean squared error and the criterion of the highest correlation coefficient between existing values and forecasted values. Research findings indicated that for all forecasting methods that have been studied the most powerful method is the combined forecasting and the forecasting model is where and represent the single forecasts at time t from Box-Jenkins method and Winters’ multiplicative exponential smoothing method respectively.