1st Edition

Online Portfolio Selection Principles and Algorithms

By Bin Li, Steven Chu Hong Hoi Copyright 2016
    230 Pages 22 B/W Illustrations
    by CRC Press

    230 Pages 22 B/W Illustrations
    by CRC Press

    With the aim to sequentially determine optimal allocations across a set of assets, Online Portfolio Selection (OLPS) has significantly reshaped the financial investment landscape. Online Portfolio Selection: Principles and Algorithms supplies a comprehensive survey of existing OLPS principles and presents a collection of innovative strategies that leverage machine learning techniques for financial investment.

    The book presents four new algorithms based on machine learning techniques that were designed by the authors, as well as a new back-test system they developed for evaluating trading strategy effectiveness. The book uses simulations with real market data to illustrate the trading strategies in action and to provide readers with the confidence to deploy the strategies themselves. The book is presented in five sections that:

    1. Introduce OLPS and formulate OLPS as a sequential decision task
    2. Present key OLPS principles, including benchmarks, follow the winner, follow the loser, pattern matching, and meta-learning
    3. Detail four innovative OLPS algorithms based on cutting-edge machine learning techniques
    4. Provide a toolbox for evaluating the OLPS algorithms and present empirical studies comparing the proposed algorithms with the state of the art
    5. Investigate possible future directions

    Complete with a back-test system that uses historical data to evaluate the performance of trading strategies, as well as MATLAB® code for the back-test systems, this book is an ideal resource for graduate students in finance, computer science, and statistics. It is also suitable for researchers and engineers interested in computational investment.

    Readers are encouraged to visit the authors’ website for updates: http://olps.stevenhoi.org.

    I: INTRODUCTION

    Introduction
    Background
    What Is Online Portfolio Selection?
    Methodology
    Book Overview

    Problem Formulation
    Problem Settings
    Transaction Costs and Margin Buying Models
    Evaluation
    Summary

    II: Principles

    Benchmarks
    Buy-and-Hold Strategy
    Best Stock Strategy
    Constant Rebalanced Portfolios

    Follow the Winner
    Universal Portfolios
    Exponential Gradient
    Follow the Leader
    Follow the Regularized Leader
    Summary

    Follow the Loser
    Mean Reversion
    Anticorrelation
    Summary

    Pattern Matching
    Sample Selection Techniques
    Portfolio Optimization Techniques
    Combinations
    Summary

    Meta-Learning
    Aggregating Algorithms
    Fast Universalization
    Online Gradient and Newton Updates
    Follow the Leading History
    Summary

    III: Algorithms

    Correlation-Driven Nonparametric Learning
    Preliminaries
    Formulations
    Algorithms
    Analysis
    Summary

    Passive–Aggressive Mean Reversion
    Preliminaries
    Formulations
    Algorithms
    Analysis
    Summary

    Confidence-Weighted Mean Reversion
    Preliminaries
    Formulations
    Algorithms
    Analysis
    Summary

    Online Moving Average Reversion
    Preliminaries
    Formulations
    Algorithms
    Analysis
    Summary

    IV: Empirical Studies

    Implementations
    The OLPS Platform
    Data
    Setups
    Performance Metrics
    Summary

    Empirical Results
    Experiment 1: Evaluation of Cumulative Wealth
    Experiment 2: Evaluation of Risk and Risk-Adjusted Return
    Experiment 3: Evaluation of Parameter Sensitivity
    Experiment 4: Evaluation of Practical Issues
    Experiment 5: Evaluation of Computational Time
    Experiment 6: Descriptive Analysis of Assets and Portfolios
    Summary

    Threats to Validity
    On Model Assumptions
    On Mean Reversion Assumptions
    On Theoretical Analysis
    On Back-Tests
    Summary

    V: Conclusion

    Conclusions
    Future Directions

    Appendix A: OLPS: A Toolbox for Online Portfolio Selection
    Introduction
    Framework and Interfaces
    Strategies
    Summary

    Appendix B: Proofs and Derivations
    Proof of CORN
    Derivations of PAMR
    Derivations of CWMR
    Derivation of OLMAR

    Appendix C: Supplementary Data and Portfolio Statistics

    Bibliography

    Index

    Biography

    Dr. Bin Li received a bachelor’s degree in computer science from Huazhong University of Science and Technology, Wuhan, China, and a bachelor’s degree in economics from Wuhan University, Wuhan, China, in 2006. He earned a PhD degree from the School of Computer Engineering of Nanyang Technological University, Singapore, in 2013. He completed the CFA Program in 2013 and is currently an associate professor of finance at the Economics and Management School of Wuhan University. Dr. Li was a postdoctoral research fellow at the Nanyang Business School of Nanyang Technological University. His research interests are computational finance and machine learning. He has published several academic papers in premier conferences and journals.

    Dr. Steven C.H. Hoi
    received his bachelor’s degree in computer science from Tsinghua University, Beijing, China, in 2002, and both his master’s and PhD degrees in computer science and engineering from The Chinese University of Hong Kong, Hong Kong, China, in 2004 and 2006, respectively. He is currently an associate professor in the School of Information Systems, Singapore Management University, Singapore. Prior to joining SMU, he was a tenured associate professor in the School of Computer Engineering, Nanyang Technological University, Singapore. His research interests are machine learning and data mining and their applications to tackle real-world big data challenges across varied domains, including computational finance, multimedia information retrieval, social media, web search and data mining, computer vision and pattern recognition, and so on.

    "Ever since access to financial data, storage capacity, and computing power stopped acting as barriers to entry, institutional-quality asset allocation solutions have become widely available to individual investors and financial advisors. Coupled with easy access to inexpensive building blocks like Exchange-Traded Funds, this dynamic has brought the spectre of digital disruption to the asset management industry. In Online Portfolio Selection, Li and Hoi do an excellent job explaining what’s actually under the hood of the "robo-advisor" applications. Unlike many books on related financial technology subjects, they don’t leave the reader with only high-level rhetoric on machine learning and financial technology, but instead roll up their sleeves and delve into the nuts and bolts of the various algorithms that power this irreversible trend. A must-read."
    —Guy Weyns, PhD., Partner, NGEN Capital, London

    "This is an excellent book showing a comprehensive menu of state-of-the-art online machine-learning algorithms in online portfolio selection and trading. It explains clearly how different algorithms can perform based on data-driven patterns that are exploited using intensive computational methods. It is a must-read for serious quantitative traders."
    Lim Kian Guan, PhD., OUB Chair Professor of Quantitative Finance, Singapore Management University