UCL Discovery
UCL home » Library Services » Electronic resources » UCL Discovery

Validation of trading strategies in the foreign exchange

Idone, Lucio; (2021) Validation of trading strategies in the foreign exchange. Doctoral thesis (Ph.D), UCL (University College London). Green open access

[thumbnail of PhD Thesis Idone.pdf]
PhD Thesis Idone.pdf

Download (10MB) | Preview


The aftermath of the recent financial crisis has caused the narrowing of investment opportunities for foreign exchange (FX) traders and investors. A debate about the profitability of trading strategies in FX has started among practitioners and academic researchers who have wondered whether is still possible to obtain positive excess returns (alpha). In this research I validate a set of trading strategies for FX. Seven experiments are carried out on macroeconomic market factors like trendfollowing, carry and value, separately. The outcome holds that the dissolution of synchronous monetary policies increases the probability of observing trends and carry opportunities in the FX. The failure of the uncovered interest rate parity by the so-called forward rate puzzle and that of the purchase parity power open opportunities for strategies like momentum, carry and value. Carry is not only applicable to spot rates as can also be used to trade FX options. Two experiments are performed to study the consistency of FX option premia and the performance of carry trade for options. For short-dated options, like the weekly ones, carry cannot produce material profits as the error implied by the forward rate is not large enough. Conversely, the premium earned from trading FX call options is a consistent source. A second line of research is dedicated to the analysis of trading strategies for FX highfrequency data. This study consists of implementing machine learning algorithms, like the exponentially-smoothing recurrent neural networks (RNN), to forecast future prices and derive a trading strategy from it. The training of these models appear to be computationally intensive but simpler than that of other neural networks like the long-short-term memory ones (LSTM). The accuracy of the forecast is adequate with no signs of over-fitting. The performance appears to be highly influenced by the presence of intra-day seasonality and jumps. A range of solutions are explored to address such a limitation.

Type: Thesis (Doctoral)
Qualification: Ph.D
Title: Validation of trading strategies in the foreign exchange
Event: UCL (University College London)
Open access status: An open access version is available from UCL Discovery
Language: English
Additional information: Copyright © The Author 2021. Original content in this thesis is licensed under the terms of the Creative Commons Attribution 4.0 International (CC BY 4.0) Licence (https://creativecommons.org/licenses/by/4.0/). Any third-party copyright material present remains the property of its respective owner(s) and is licensed under its existing terms. Access may initially be restricted at the author’s request.
UCL classification: UCL
UCL > Provost and Vice Provost Offices > UCL BEAMS
UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science
URI: https://discovery.ucl.ac.uk/id/eprint/10133586
Downloads since deposit
Download activity - last month
Download activity - last 12 months
Downloads by country - last 12 months

Archive Staff Only

View Item View Item