A hybrid technique for dinar coin price prediction using artificial neural network based autogressive modeling technique
The recent introduction of Islamic Gold (Dinar) and Silver (Dirham) coins around the world has brought about a new paradigm in the world financial, economic and monetary system. The importance of accurately predicting the price of these coins ahead of time will contribute significantly to its usage...
| Main Authors: | , , |
|---|---|
| Format: | Conference or Workshop Item |
| Language: | English |
| Published: |
2011
|
| Subjects: | |
| Online Access: | http://irep.iium.edu.my/1767/ http://irep.iium.edu.my/1767/ http://irep.iium.edu.my/1767/1/A_hybrid_technique_for_dinar.pdf |
| Summary: | The recent introduction of Islamic Gold (Dinar) and Silver (Dirham) coins around the world has brought about a new paradigm in the world financial, economic and monetary system. The importance of accurately predicting the price of these coins ahead of time will contribute significantly to its usage for: daily transaction, investment and development of necessary infrastructures for the universal adoption of these coins. Thus in this work, recently proposed artificial neural network based autoregressive (ANN-BASED AR) modeling technique has been applied in predicting accurately the daily price of Islamic Dinar coin. The input data is formatted to meet the input data requirement of the ANN-based AR model. The formatted data are then fed to the ANN-based AR model for parameters estimation. Upon convergence, the required model coefficients are computed from the synaptic weights and adaptive coefficients of the activated function in a two layer feed forward back-propagation artificial neural network (ANN) system. Performance analysis of the proposed approach shows that this proposed hybrid technique can accurately predict the price of Dinar coin and the use of this approach shows better performance when compared to the use of linear prediction technique. Other likely areas of application of the proposed approach have also been presented in this paper. |
|---|