Hansun, Seng and Kristanda, Marcel Bonar (2017) Performance Analysis of Conventional Moving Average Methods in Forex Forecasting. In: Proceedings of 2017 International Conference on Smart Cities, Automation & Intelligent Computing Systems (ICON-SONICS 2017), 08 November 207, Yogyakarta.
Full text not available from this repository.Abstract
Time series forecasting is one of the main goals of time series analysis, whether it uses conventional, advance, or hybrid method. One of the most popular and commonly used methods is the moving average (MA) method, which comes with so many variations. Some of the basic and well-known MA methods are simple moving average (SMA), weighted moving average (WMA), and exponential moving average (EMA). In this study, we would like to do a performance analysis of those methods, especially in Forex transaction data. Three major currency pairs been used in this research are EUR/USD, AUD/USD, and GBP/USD, with a total of 1,287 records. From the experimental results taken, we have EMA as the best MA method, followed by WMA and SMA consecutively. It is shown by using mean square error (MSE), mean absolute percentage error (MAPE), and mean absolute scaled error (MASE) that EMA has the smallest average values for all three forecast error measurements, i.e. 0.000051927 for MSE, 0.44720 for MAPE, and 1.07697 for MASE.
| Item Type: | Conference or Workshop Item (Paper) |
|---|---|
| Keywords: | Forex forecasting; SMA; WMA; EMA; performance analysis |
| Subjects: | 000 Computer Science, Information and General Works > 000 Computer Science, Knowledge and Systems > 004 Computer Science, Data Processing, Hardware > 004.2 Systems Analysis and Design, Information Architecture, Performance Evaluation 300 Social Sciences > 330 Economics > 332 Financial Economics (Shares, Investment) |
| Divisions: | Universitas Multimedia Nusantara |
| Depositing User: | Administrator UMN Library |
| Date Deposited: | 23 Feb 2018 08:37 |
| Last Modified: | 24 Jan 2023 02:20 |
| URI: | https://kc.umn.ac.id/id/eprint/2774 |
Actions (login required)
 |
View Item |
