University of Twente Student Theses
WM/Reuters fixing rate prediction for algorithmic FX trading using machine learning
Timmerman, L.D. (2022) WM/Reuters fixing rate prediction for algorithmic FX trading using machine learning.
|
PDF
5MB |
| Abstract: | When investing in assets that are traded in a foreign currency, pension funds are exposed to currency risks. The asset managers of the pension funds can use different techniques to hedge themselves against this risk. As a part of this protection, the managers execute a significant amount of foreign exchange (FX) spot transactions. Within the FX spot market, time window, 15:57:30 to 16:02:30 UTC, is called the WMR 4 pm fixing window and is used for fixing orders and as a benchmark for numerous reference purposes. This thesis aims to develop a deep learning model that can be used to predict the WMR of the EUR/USD, one day in advance, by finding patterns in FX datasets and market data analysis. This model can provide traders with a prediction of future movements in the market, which can then be used to optimise the execution strategy or execution moment, given the FX orders that need to be traded. The best performing deep learning model is an CNN-LSTM architecture. It outperforms more standard regression models, other deep learning methods, and guessing, for both direction and size of the movement. However, it does lack in consistent convergence and explanatory power. |
| Item Type: | Essay (Master) |
| Clients: | MN, Netherlands PGGM, Netherlands |
| Faculty: | BMS: Behavioural, Management and Social Sciences |
| Subject: | 54 computer science, 83 economics |
| Programme: | Industrial Engineering and Management MSc (60029) |
| Link to this item: | https://purl.utwente.nl/essays/90471 |
| Export this item as: | BibTeX EndNote HTML Citation Reference Manager |
Show download statistics for this publication
Show download statistics for this publicationRepository Staff Only: item control page