Table of Contents

Portfolio Optimization in the Context of Cointelated Pairs: Stochastic Differential Equation vs. Machine Learning Approach

Babak Mahdavi-Damghani, University of Oxford
Konul Mustafayeva, King's College London
Stephen Roberts, University of Oxford - Oxford-Man Institute of Quantitative Finance
Cristin Buescu, King's College London, Department of Mathematics


FINANCIAL ENGINEERING eJOURNAL

"Portfolio Optimization in the Context of Cointelated Pairs: Stochastic Differential Equation vs. Machine Learning Approach" Free Download

BABAK MAHDAVI-DAMGHANI, University of Oxford
Email:
KONUL MUSTAFAYEVA, King's College London
Email:
STEPHEN ROBERTS, University of Oxford - Oxford-Man Institute of Quantitative Finance
Email:
CRISTIN BUESCU, King's College London, Department of Mathematics
Email:

The cointelation model recently introduced is studied in the context of portfolio optimization. The proposed models are twofold. The first model studied takes the Stochastic Differential Equation approach in which the methodology is split in a dual switching exercise using some of the classic results of Markowitz Modern portfolio theory with an Ornstein-Uhlenbeck optimized overlay. The methodology is then compared to what is presented as the Machine Learning mirror methodology in the form of the new Band-wise Gaussian Mixture model which we expose as giving similar results while keeping the methodology simpler and more adaptable to regime change.

^top

About this eJournal

This eJournal distributes working and accepted paper abstracts related to development and employment of quantitative techniques to further our understanding of financial markets, instruments, and strategies. The eJournal welcomes research with a focus on advancing the theory or practice of financial engineering in endowments, hedge funds, insurance firms, investment and commercial banks, pension funds, and personal financial and retirement planning. Topics of interest include, but are not limited to, econometric analysis of financial data, enterprise risk management, investment and consumption models, optimal portfolio, pricing and hedging of financial instruments, as well as innovative empirical studies, analytical models, and mathematical algorithms in credit, energy, fixed-income and other markets.

Submissions

To submit your research to SSRN, sign in to the SSRN User HeadQuarters, click the My Papers link on left menu and then the Start New Submission button at top of page.

Distribution Services

If your organization is interested in increasing readership for its research by starting a Research Paper Series, or sponsoring a Subject Matter eJournal, please email: RPS@SSRN.com

Distributed by

Management Research Network (MRN), a division of Social Science Electronic Publishing (SSEP) and Social Science Research Network (SSRN)

Directors

OPER SUBJECT MATTER EJOURNALS

MICHAEL C. JENSEN
SSRN, Harvard Business School, National Bureau of Economic Research (NBER), European Corporate Governance Institute (ECGI), Harvard University - Accounting & Control Unit
Email: mjensen@hbs.edu

Please contact us at the above addresses with your comments, questions or suggestions for OPER-Sub.