Pomorski, Piotr;
(2024)
Construction of Effective Regime-Switching Portfolios Using a Combination of Machine Learning and Traditional Approaches.
Doctoral thesis (Ph.D), UCL (University College London).
Preview |
Text
Thesis_Piotr_Pomorski_Final.pdf - Accepted Version Download (4MB) | Preview |
Abstract
This thesis proposes and tests a detection-prediction-optimisation regimeswitching framework to harness the regime shifts in financial time series for the advantage of portfolio managers. The work is divided into three research objectives, each explored in a separate chapter. First, however, Chapter 2 provides a thorough review of relevant literature and lays the foundation for the subsequent research chapters. Chapter 3 then focuses on the implementation of a novel regime detection framework that combines a technical indicator with a statistical method of regime detection. The resulting KAMA+MSR model outperforms the benchmarks in terms of stability and accuracy. Chapter 4 moves from detection to prediction, exploring the prediction of financial regimes ex-ante, using the regime labels from Chapter 3 and a Random Forest model as a predictor. Models for three different asset classes show solid out-of-sample classification performance and achieve excellent financial results (as measured by the Sortino ratio) based on a long/short trading strategy. Chapter 5 addresses the contributions and limitations of Chapter 4 by focusing on the third research objective: the construction of realistic, regime-robust portfolios. The constructed portfolio from Chapter 5 outperforms its benchmarks, despite not allowing shorting positions due to potential ethical and institutional constraints. The combined contributions of these three research chapters could serve as key components of a quantitative trading system, emphasising the practical aspect of this research.
| Type: | Thesis (Doctoral) |
|---|---|
| Qualification: | Ph.D |
| Title: | Construction of Effective Regime-Switching Portfolios Using a Combination of Machine Learning and Traditional Approaches |
| Open access status: | An open access version is available from UCL Discovery |
| Language: | English |
| Additional information: | Copyright © The Author 2024. Original content in this thesis is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0) Licence (https://creativecommons.org/licenses/by-nc/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 UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science > Dept of Computer Science |
| URI: | https://discovery.ucl.ac.uk/id/eprint/10192012 |
Loading...
Loading...Loading...Loading...Archive Staff Only
 |
View Item |
