A hybrid forecasting model for Malaysian exports earning of natural rubber-based products

• Main Authors: Sallehuddin, Roselina, Shamsuddin, Siti Mariyam, Talib, Mohd. Shukor
• Format: Conference or Workshop Item
• Language: English
• Published: 2006
• Subjects: Q Science (General)
• Online Access: http://eprints.utm.my/id/eprint/3183/1/B-03_Roselina_Bandung.pdf; http://eprints.utm.my/id/eprint/3183/
• Institution: Universiti Teknologi Malaysia

Natural rubber is one of the most important crops in Malaysia. Malaysia is the third largest producer of rubber in the world after Thailand and Indonesia. As Malaysia’s economic shifted from primary industries to manufacturing, Malaysia becomes one of the world biggest exporters of natural rubber-based products (NRP) such as gloves, catheters and latex threads. One of the major goals of agricultural policy in Malaysia is to increase foreign exchange earning through increasing exports of natural rubber-based products. However, fluctuation in the export earning will lead to the uncertainty in the level of foreign exchange earning. For example, in 2003, 2004 and 2005 the rubber product industry contributed RM 6.07, RM 7.88 and RM8.03 billion respectively to the country export earnings. Therefore, the aim of this study is to propose a hybrid forecasting model to predict the export earning for natural rubber based product. This hybrid model will consist of Multiple Regression (MR) and Artificial Neural Network (ANN). Several factors that will affect the export earnings such as the production of natural rubber-based product, the price of natural rubber and the price of synthetic rubber also will be considered. To benchmark the result produced by the propose model, comparison with traditional techniques (Multiple Regression) will be conducted.