File Download
There are no files associated with this item.
Links for fulltext
(May Require Subscription)
- Publisher Website: 10.1109/IC3I46837.2019.9055674
- Scopus: eid_2-s2.0-85083571904
Supplementary
-
Citations:
- Scopus: 0
- Appears in Collections:
Conference Paper: Genetic Algorithm Based Bidirectional Recurrent Neural Network for LIBOR Prediction
Title | Genetic Algorithm Based Bidirectional Recurrent Neural Network for LIBOR Prediction |
---|---|
Authors | |
Keywords | Machine learning Libor genetic algorithm bidirectional recurrent neural network |
Issue Date | 2020 |
Publisher | IEEE. |
Citation | 2019 International Conference on Contemporary Computing and Informatics (IC3I), Singapore, 12-14 December 2019, p. 9-15 How to Cite? |
Abstract | London Inter Banking Offered Rate (“LIBOR”) is one of the most important indicators of global currency liquidity. LIBOR market is the most important interest rate market with huge market volume and millions market stake holders from different countries trade in the market with different currencies. Because of the complexity of the stakeholders and the currencies, it is difficult to forecast the LIBOR which is important for the stability of global finance market. To train and predict the time series dataset, the traditional algorithms, including pure genetic algorithm and recurrent neural network, have the performance issues when forecasting the turning point in a one-direction trend. This paper is to propose a new algorithm, which is to combine genetic algorithm with bidirectional recurrent neural network (GA+BRNN) to improve the performance of LIBOR prediction. The pro and cons of the new algorithm will be introduced with the evaluation criteria including Mean Squared Error (“MSE”) and Fitness Value. 50 test cases are executed randomly in the experiment phase to display the performance of prediction for Libor in USD in different test scenarios. The target variance between predication and real value is no more than 0.015. |
Persistent Identifier | http://hdl.handle.net/10722/289849 |
ISBN |
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Tan, X | - |
dc.date.accessioned | 2020-10-22T08:18:21Z | - |
dc.date.available | 2020-10-22T08:18:21Z | - |
dc.date.issued | 2020 | - |
dc.identifier.citation | 2019 International Conference on Contemporary Computing and Informatics (IC3I), Singapore, 12-14 December 2019, p. 9-15 | - |
dc.identifier.isbn | 9781728155302 | - |
dc.identifier.uri | http://hdl.handle.net/10722/289849 | - |
dc.description.abstract | London Inter Banking Offered Rate (“LIBOR”) is one of the most important indicators of global currency liquidity. LIBOR market is the most important interest rate market with huge market volume and millions market stake holders from different countries trade in the market with different currencies. Because of the complexity of the stakeholders and the currencies, it is difficult to forecast the LIBOR which is important for the stability of global finance market. To train and predict the time series dataset, the traditional algorithms, including pure genetic algorithm and recurrent neural network, have the performance issues when forecasting the turning point in a one-direction trend. This paper is to propose a new algorithm, which is to combine genetic algorithm with bidirectional recurrent neural network (GA+BRNN) to improve the performance of LIBOR prediction. The pro and cons of the new algorithm will be introduced with the evaluation criteria including Mean Squared Error (“MSE”) and Fitness Value. 50 test cases are executed randomly in the experiment phase to display the performance of prediction for Libor in USD in different test scenarios. The target variance between predication and real value is no more than 0.015. | - |
dc.language | eng | - |
dc.publisher | IEEE. | - |
dc.relation.ispartof | International Conference on Contemporary Computing and Informatics (IC3I) 2019 | - |
dc.rights | International Conference on Contemporary Computing and Informatics (IC3I) 2019. Copyright © IEEE. | - |
dc.rights | ©2020 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. | - |
dc.subject | Machine learning | - |
dc.subject | Libor | - |
dc.subject | genetic algorithm | - |
dc.subject | bidirectional recurrent neural network | - |
dc.title | Genetic Algorithm Based Bidirectional Recurrent Neural Network for LIBOR Prediction | - |
dc.type | Conference_Paper | - |
dc.description.nature | link_to_subscribed_fulltext | - |
dc.identifier.doi | 10.1109/IC3I46837.2019.9055674 | - |
dc.identifier.scopus | eid_2-s2.0-85083571904 | - |
dc.identifier.hkuros | 317076 | - |
dc.identifier.spage | 9 | - |
dc.identifier.epage | 15 | - |
dc.publisher.place | Singapore | - |