A Forecasting Model for the Export Volume of Frozen Fish Fillets in Thailand

Abstract

The purpose of this study is to develop an appropriate forecasting model for the export volume of frozen fish fillets in Thailand using four statistical methods: Holt’s exponential smoothing, Brown’s exponential smoothing, and the Damped trend exponential smoothing method. The time series data on frozen fish meat export volume were gathered from the Office of Agricultural Economics website from January 2018 to November 2024. The 83 observations were divided into two datasets. The first dataset had 77 observations from January 2018 to May 2024. The second dataset had 6 observations from June 2024 to November 2024. The first dataset for creating the forecasting model. The second dataset is for checking the accuracy of the forecasting models via the criterion of the lowest mean absolute percentage error (MAPE) and the lowest root mean square error (RMSE), with the model yielding the lowest values considered the most suitable. The results showed that Holt’s exponential smoothing method is most suitable for this time series (MAPE = 21.650 and RMSE = 1267.954).