Mohanaeswari Kannan (2009) Presences of missing valves in share market forecasting. Universiti Malaysia Sabah. (Unpublished)
|
Text
ae0000003014.pdf Download (8MB) | Preview |
Abstract
This dissertation is on observing the influence of the missing values towards the prediction of NSTP (M) Berhad share price. The opening and the closing price of NSTP (M) Berhad will be forecasted starting from 1st September 2008 till 31st December 2008. The missing values in this study are estimated using the cubic spline interpolation using C programming. The stationarity of each series is achieved using transformation before the elimination is carried out. Since the ARMA with P + q ≤ 10 is used, 65 possible models are generated and by using the Eviews software elimination is done based on the p value. Eight selection criteria (8SC) are used to choose the best model out of the selected for each series. From the selection using 8SC the series without missing values for both opening and closing price is selected as the best models. The forecasting is done using both best models and the accuracy is compared. Besides checking the accuracy of the forecasting, the Mean Absolute Percentage Error (MAPE) is used to check the time period that this best model can be used to forecast. The closing price is known to give a better forecast compared to the opening. Also found that, the forecasting using this ARMA best model is applicable for short period prediction.
| Item Type: | Academic Exercise |
|---|---|
| Keyword: | stationarity, missing values, forecasting, ARMA, 8Se, Eviews |
| Subjects: | Q Science > QA Mathematics |
| Department: | SCHOOL > School of Science and Technology |
| Depositing User: | ADMIN ADMIN |
| Date Deposited: | 03 Aug 2016 14:16 |
| Last Modified: | 20 Nov 2017 11:40 |
| URI: | https://eprints.ums.edu.my/id/eprint/13663 |
Actions (login required)
 |
View Item |