Double exponential smoothing and Holt-Winters methods with optimal initial values and weighting factors for forecasting lime, Thai chili and lemongrass prices in Thailand

Abstract

In this paper, the Double Exponential Smoothing (DES), the Multiplicative Holt-Winters (MHW) and the Additive Holt-Winters (AHW) methods with optimal initial values and weighting factors are presented to forecast lime, Thai chili, and lemongrass prices in Thailand. Since these plants are important economic plants in Thailand, knowing their market prices or trends before selling them would be very useful for Thai agriculturists to appropriately plan their work and sales. In this paper, lime, Thai chili and lemongrass prices from January 2011 to September 2016 were gathered from the website’s database of Simummuang market used and as input data. This is one of the biggest markets in Thailand. Our study reveals that the DES method with optimal initial values and weighting factors provides the smallest forecasting error measured by the Mean Absolute Percentage Error (MAPE) when forecasting Thai chili and lemongrass prices, while the MHW and the AHW methods show better performance on forecasting lime price data which present seasonality patterns. The forecasted prices of lime, Thai chili and lemongrass during October 2016 to December 2016 are also provided.

Keywords

• Forecasting
• Double exponential smoothing
• Holt-Winters
• Lime
• Thai chili
• Lemongrass