Tiew, Yee Wen (2012) Development of forecasting model for sawn timber export price in Sabah. Universiti Malaysia Sabah. (Unpublished)
|
Text
ae0000002924.pdf Download (10MB) | Preview |
Abstract
Sabah timber industry is facing a shortage of raw material due to past over-logging and lack of sustainable forest management program (Borneo Post Online, 2012). This critical situation has forced this industry to depend on other means for sustainability, especially for sawn timber. Hence, this research endeavours to give a mathematical approach for a more sustainable industry by modelling the export price of sawn timber in Sabah. lime series and artificial neural network (ANN) are used to develop forecasting model of the sawn timber export price. In this research, Autoregressive Moving Average (ARMA) model is utilized which incorporates the times series model building procedures, while ANN is solved using the MATLAB Toolbox with one dependent and two independent variables. The historical monthly data on sawn timber export price of 228 observations is acquired from the Department Statistic of Sabah from 1991 to 2009. The data must achieve the assumption of stationary before the ARMA model is determined. The best model in ARMA and ANN is determined by obtaining the minimum value in the eight selection criteria (SSC). Finally, MAPE is used to check the validity of the best model. The results show that the best model for ARMA model is M23.0.1, while best neural network model is the single layer of 4 which is based on the R2 value, MSE value and standardized residuals.
| Item Type: | Academic Exercise |
|---|---|
| Keyword: | export price, sawn timber, time series, artificial neural network (ANN), forecasting model, Autoregressive Moving Average (ARMA) model |
| Subjects: | Q Science > QA Mathematics |
| Department: | SCHOOL > School of Science and Technology |
| Depositing User: | ADMIN ADMIN |
| Date Deposited: | 30 Jun 2016 09:53 |
| Last Modified: | 07 Nov 2017 11:13 |
| URI: | https://eprints.ums.edu.my/id/eprint/13544 |
Actions (login required)
 |
View Item |