Multiple regression model for forecasting quantity of supply of off-season longan

• Main Authors: Chompoonoot Kasemset, Nisachon Sae-Haew, Apichat Sopadang
• Format: Journal
• Published: 2018
• Subjects: Multidisciplinary
• Online Access: https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=84963955004&origin=inward; http://cmuir.cmu.ac.th/jspui/handle/6653943832/53937
• Institution: Chiang Mai University

This research work aims to develop a forecasting model to predict the quantity of supply of off-season longan using multiple regression technique. There are 23 factors that influence the quantity of supply of off-season longan. Data collection was done in Chiang Mai and Lamphun provinces. The forecasting model based on multiple regression techniques, with enter, forward, backward, and stepwise selection methods were adopted, and these methods yielded mean absolute percentage error (MAPE) values of 18.39%, 25.63%, 21.21%, and 25.63%, respectively. These results demonstrate that multiple regression with the enter selection method is practical to predict the quantity of supply of off-season longan.