Market Momentum E-Book

Stephen Satchell

• Herausgeber: John Wiley & Sons
• Kategorie: Fachliteratur
• Serie: Wiley Finance Series
• Sprache: Englisch

Seitenzahl: 850

Veröffentlichungsjahr: 2020

Beschreibung

A one-of-a-kind reference guide covering the behavioral and statistical explanations for market momentum and the implementation of momentum trading strategies Market Momentum: Theory and Practice is a thorough, how-to reference guide for a full range of financial professionals and students. It examines the behavioral and statistical causes of market momentum while also exploring the practical side of implementing related strategies. The phenomenon of momentum in finance occurs when past high returns are followed by subsequent high returns, and past low returns are followed by subsequent low returns. Market Momentum provides a detailed introduction to the financial topic, while examining existing literature. Recent academic and practitioner research is included, offering a more up-to-date perspective. What type of book is Market Momentum and how does it serve a range of readers' interests and needs? * A holistic market momentum guide for industry professionals, asset managers, risk managers, firm managers, plus hedge fund and commodity trading advisors * Advanced text to help graduate students in finance, economics, and mathematics further develop their funds management skills * Useful resource for financial practitioners who want to implement momentum trading strategies * Reference book providing behavioral and statistical explanations for market momentum Due to claims that the phenomenon of momentum goes against the Efficient Markets Hypothesis, behavioral economists have studied the topic in-depth. However, many books published on the subject are written to provide advice on how to make money. In contrast, Market Momentum offers a comprehensive approach to the topic, which makes it a valuable resource for both investment professionals and higher-level finance students. The contributors address momentum theory and practice, while also offering trading strategies that practitioners can study.

Leseprobe

Market Momentum

Theory and Practice

By

STEPHEN SATCHELL

and

ANDREW GRANT

This edition first published 2021

© 2021 Stephen Satchell and Andrew Grant

Registered office

John Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, United Kingdom

For details of our global editorial offices, for customer services and for information about how to apply for permission to reuse the copyright material in this book please see our website at www.wiley.com.

All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, except as permitted by the UK Copyright, Designs and Patents Act 1988, without the prior permission of the publisher.

Wiley publishes in a variety of print and electronic formats and by print-on-demand. Some material included with standard print versions of this book may not be included in e-books or in print-on-demand. If this book refers to media such as a CD or DVD that is not included in the version you purchased, you may download this material at http://booksupport.wiley.com. For more information about Wiley products, visit www.wiley.com.

Designations used by companies to distinguish their products are often claimed as trademarks. All brand names and product names used in this book are trade names, service marks, trademarks or registered trademarks of their respective owners. The publisher is not associated with any product or vendor mentioned in this book.

Limit of Liability/Disclaimer of Warranty: While the publisher and author have used their best efforts in preparing this book, they make no representations or warranties with respect to the accuracy or completeness of the contents of this book and specifically disclaim any implied warranties of merchantability or fitness for a particular purpose. It is sold on the understanding that the publisher is not engaged in rendering professional services and neither the publisher nor the author shall be liable for damages arising herefrom. If professional advice or other expert assistance is required, the services of a competent professional should be sought.

Library of Congress Cataloging-in-Publication Data

Names: Grant, Andrew Robert, 1982- author. | Satchell, Stephen Ellwood, 1949- author.

Title: Market momentum : theory and practice / Andrew Robert Grant, Stephen Ellwood Satchell.

Description: First Edition. | Hoboken : Wiley, 2020. | Series: The wiley finance series | Includes index.

Identifiers: LCCN 2020020406 (print) | LCCN 2020020407 (ebook) | ISBN 9781119599326 (hardback) | ISBN 9781119599470 (adobe pdf) | ISBN 9781119599371 (epub)

Subjects: LCSH: Investment analysis. | Securities—Prices. | Economics—Psychological aspects.

Classification: LCC HG4529 .G73 2020 (print) | LCC HG4529 (ebook) | DDC 332.63/2042—dc23

LC record available at https://lccn.loc.gov/2020020406

LC ebook record available at https://lccn.loc.gov/2020020407

Cover Design: Wiley

Cover Image: ©palamatic/shutterstock

Abstracts

1 Andrew Grant

This chapter examines the behavioural finance argument for the existence of momentum profits. From a behavioural finance perspective, asset prices may deviate from fundamental values, which can persist if market frictions prevent a prompt correction to mispricing. As risk, in the form of a Fama-French three-factor model, has been shown to provide a poor explanation for momentum returns, academics have sought psychology-inspired reasons for the phenomenon. We review the literature on behavioural finance and momentum, starting with the theoretical studies of the late 1990s, which have become highly influential. Following on from this, we discuss the recent empirical evidence supporting predictions such as slow information diffusion, incorrect updating of beliefs, trading at the 52-week high, individual investor trading and market-wide sentiment. Among the key insights is that behavioural finance can help provide an explanation for statistical patterns that generate momentum portfolios (as in Chapter 2) and may help practitioners in identifying themes for enhancing their investment portfolios.

2 Steve Satchell

In this chapter we look at different momentum strategies and their properties. Many of the empirical results of momentum strategies can be seen to result from the structure of return processes and the design of the strategy. In particular, considering simple cases, we investigate the return distributions of what are the major momentum strategies which are cross-sectional momentum (CSM), time-series momentum (TSM), relative strength strategies (RSS) and cross-asset momentum. For the case of two assets, we can say a great deal about the structure and properties of returns. From a statistical perspective, behavioural analysis will determine the magnitude of model parameters along with, in some cases, the specification of the model itself. Taking these as given, the statistical analysis will determine the properties of the strategy returns. In this form of analysis, we are interested in the population moments such as the mean, variance, skewness and kurtosis but also the time-series moments of the momentum process. It is hoped that the relatively simple analysis in this chapter will help the reader in later chapters.

3 Nick Baltas and Robert Kosowski

Motivated by studies of the impact of frictions on asset prices, we examine the effect of key components of time-series momentum strategies on turnover and performance. We show that more efficient volatility estimation and price-trend detection can significantly reduce portfolio turnover by more than one-third, without causing a statistically significant performance degradation. We propose a novel implementation of the strategy that incorporates the pairwise signed correlations by means of a dynamic leverage mechanism. The correlation-adjusted variant outperforms the naïve implementation of the strategy and the outperformance is more pronounced in the post-2008 period. Finally, using a transaction costs model for futures-based strategies that separates costs into roll-over and rebalancing costs, we show that our findings remain robust to the inclusion of transaction costs

4 Jose Menchero and Lei Ji

In this chapter, we study the risk and return of momentum in developed equity markets. We construct factor portfolios by cross-sectional regression. Univariate regression results in ‘simple’ factor portfolios that contain ‘incidental’ bets on other factors. Multivariate regression results in ‘pure’ factor portfolios that are neutral to all factors except momentum. We compare performance of simple and pure momentum factors across various developed markets. We find that simple and pure factors have virtually identical long-term performance within each equity market. The pure factors, however, achieve this performance with considerably lower volatility, resulting in higher risk-adjusted performance. We also study the volatilities and correlations of momentum factors across time. We find that these quantities peaked during the Internet Bubble and the Financial Crisis. Finally, we show that for most periods, momentum has been negatively correlated with the market, thus offering attractive diversification opportunities.

5 Dan diBartolomeo and Bill Zieff

Price momentum exposure is a familiar concept to anyone who has studied equity return factors. Less addressed in the literature is how momentum effects in one asset class manifest in other asset classes. Theories of corporate finance that link equity and bond returns through their common dependence on the value of the firm have been known for some time, but these theories do not address the dynamics between debt and equity in a time-series context. Cross-Asset momentum provides a parsimonious description of such dynamics. We describe theoretical underpinnings and empirical results for several cross-asset class momentum effects. Included is a discussion of equity momentum impact into fixed income, selected momentum effects in illiquid assets, such as real estate and private equity, and momentum results pertaining to commodities. Implications for active management in asset allocation and active management within asset classes are explored and highlighted.

6 Katharina Schwaiger and Muhammad Massood

Momentum as an investment strategy is a well-known component of systematic strategies. With the rise of exchange-traded funds (ETFs) and smart beta, momentum strategies are available to a broad range of clients including retail via a fully transparent rules-based index strategy. The indices are built to weight and select stocks from an equity universe with high momentum exposure or high past returns. Those ETFs aim to offer momentum factor exposure at a competitive fee. In this chapter we shed light on the momentum ETF landscape and discuss how much momentum is in momentum ETFs. We find that momentum indices exhibit momentum characteristics despite differences in the underlying index methodologies, lack of leverage and infrequent rebalancing.

7 Oliver Williams

This chapter reviews time-series momentum (TSM) strategies as commonly used by Commodity Trading Advisers (CTAs). We compare TSM with cross-sectional momentum (CSM) and describe examples of various methods used for TSM signal generation, showing that there are certain essential similarities in TSM approaches although the degree of technical complexity varies considerably across methods. We note various stylized facts of CTA investing (e.g. relatively low Sharpe ratio and positive skewness) and highlight practically relevant results from existing literature concerning characteristics of TSM returns. Throughout the chapter we use simple market models to illustrate our points and we derive an expression for the correlation between TSM and CSM returns in this setting, which may be useful when contemplating asset allocation between these strategies.

8 Anders Petterson and Oliver Williams

In this chapter we apply momentum returns models in a non-traditional setting: the valuation of contemporary art. Specifically, we consider whether a momentum model is useful at predicting auction outcomes based on changes in subjective valuation opinions observed over a short period prior to the auction. We use a novel data set (supplied by ArtForecaster.com) that contains multiple valuations of the same piece of art provided by individual forecasters at approximately the same time. Combined with auctioneers' estimates and hammer prices this data can be used to estimate a simple dynamic model of valuation changes. We fail to find significant evidence of time-series momentum in these valuation changes, but a forecasting strategy based on cross-sectional momentum appears to be effective. This implies that subjective opinion is more informative when considered on a relative basis (one work of art versus another) rather than outright, and we conjecture this may be due to a sub-conscious bias on the part of art market participants towards positive rather than negative sentiment. We conclude that momentum models can provide a useful framework for analysis of information and behaviour outside traditional financial markets.

9 Yang Gao

This chapter explores the interaction between momentum and volatility. The literature has shown that momentum strategies using volatility as the conditioning variable strongly outperform plain momentum strategies by reporting higher alphas, increased Sharpe ratios and improved skewness. These volatility-managed momentum strategies benefit from the use of volatility, which has the potential to better forecast future returns. We empirically review the various volatility-managed momentum strategies that are proposed in the academic literature. Our findings show strong outperformance of both the time-series (TSVM) and cross-sectional volatility-managed momentum strategies (CSVM) in the USA, Europe and Australia. We document interesting evidence that the TSVM strategy seems to work in the Chinese and Japanese markets where traditional momentum or CSVM does not work well. We further discuss the role of leverage constraints, turnover and short-selling constraints in the practical use of volatility-managed momentum strategies. Last, we examine how volatility manages momentum risk and suggest that momentum is not only related to its own volatility.

10 Andrew Grant, Oh Kang Kwon and Steve Satchell

The selection of assets based on the rankings of one or more of their attributes over a prior period is a standard practice in the construction of portfolios in both academic and practitioner finance. Momentum is a familiar example of this activity. Although the properties of returns from these portfolios have been the subject of considerable empirical research, there is only limited literature investigating their theoretical properties. In this chapter, we address this gap by deriving analytically the distributional properties of these portfolio returns under the assumption that asset attributes and returns are jointly multivariate normal. We show that prior sorting of asset attributes induces non-normality in the portfolio returns and that the returns depend on the order and nature of sorting The analysis looks at two fundamental types of sorting, both commonly used by practitioners, and the authors show that the method of sorting influences the pattern of returns under commonly encountered circumstances.

11 Stefano Cavaglia, Vadim Moroz and Louis Scott

The research presented in this chapter extends and corroborates the findings of Scott and Cavaglia (2016) who first illustrated the potential benefits of a factor premia overlay to an inter-temporal wealth accumulation strategy that is fully invested in equities. We examine a panel of 5 local factor premia (inclusive of momentum) in each of 21 developed equity markets and 5 regions. In nearly all instances we find that wealth accumulation is significantly enhanced by a time invariant, equal weighted allocation to these premia. The enhancement is driven in part by the mean return of the premia but more importantly by their generally positive payoff in adverse market environments. The quality and momentum factor premia are the largest contributors to this result. Applying conventional measures of risk aversion we quantify the importance of downside protection to supporting the attainment of investors' goals; the utility-based evidence suggests that some factor premia are valued by investors even when their expected return is zero.

12 Christopher Polk, Mo Haghbin and Alessio de Longis

Factor cyclicality can be understood in the context of factor sensitivity to aggregate cash-flow news. Factors exhibit different sensitivities to macroeconomic risk, and this heterogeneity can be exploited to motivate dynamic rotation strategies among established factors: size, value, quality, low volatility and momentum. A timely and realistic identification of business cycle regimes, using leading economic indicators and global risk appetite, can be used to construct long-only factor rotation strategies with information ratios nearly 70% higher than static multifactor strategies. The rules introduced here are simple to apply and can be interpreted as a macroeconomic regime-switching model without the cumbersome technology that such models usually involve. Hidden regimes are absent and, in this framework, we know what regime we are in. Results are statistically and economically significant across regions and market segments, also after accounting for transaction costs, capacity and turnover.

13 Ron Bird, Xiaojun (Kevin) Gao and Danny Yeung

Momentum is one of numerous market anomalies highlighted over the last 30 or more years, bringing into question the efficiency of markets around the world. However, the question is left open as to whether the apparent profit opportunities offered by time-series and cross-sectional investment strategies are exploitable. With the incorporation of transaction costs and risk, the ‘typical’ implementation of both momentum strategies yields profits in only a handful of markets. However, we demonstrate that the performance of the momentum portfolios is very much dependent on the extent to which implementation rules chosen are in synchronicity with the periodicity of the market.

14 Shivam Ghosh, Steve Satchell and Nandini Srivastava

Artificial Intelligence (AI) and Machine Learning (ML) has seen unprecedented growth over the past decade with applications across healthcare, robotics, data security and automotive industries to name a few. Penetration of AI in finance has been slower; primarily due to the intrinsic nature of markets – non-stationarity of financial processes and lack of sufficient training data. We make a foray into applying ML to finance by training a range of models to trade Momentum – a systematic strategy that benefits from persistence of trends in markets. Our findings suggest that a class of ML algorithms like Random Forests and Neural Nets produce Sharpe ratios close to optimal vanilla time-series momentum (TSMOM). However, correlation between ML and vanilla TSMOM signals is low allowing us to harvest fruits of diversification by creating hybrid vanilla – ML TSMOM strategies. We find such hybrid portfolio Sharpe ratios to be twice as high as vanilla TSMOM.

15 Byoung-Kyu Min

One of the important linkages in financial economics is between the world of macroeconomics and the returns of individual companies. Whilst this is usually hard to identify empirically, it is nevertheless possible to find causality between macroeconomics and strategy returns. Furthermore, such causal relationships are of great interest to practitioners and constitute a key part of what is described as active quantitative fund management.

Recent empirical literature shows that time variation in the profitability of momentum strategies is critical to understanding the source of momentum. This chapter presents direct evidence that momentum strategies deliver significantly negative profits during ‘bad’ economic states in which investors demand the highest market risk premium, suggesting that momentum strategies expose investors to greater downside risk. It also reviews the literature investing whether momentum profits are related to various economic states, including business cycles, market state and investor sentiment, and discusses explanations for these findings.

16 Chris Tinker

Momentum has been considered a market anomaly and treated as a behavioural phenomenon. While generally regarded as a standalone trading strategy as a consequence, growing interest in factor-based investing and the emergence of Alternative Risk Premia (ARP), argues for a more integrated, investment-based analysis of momentum. Under a systematic, stock level, time-series forecast framework we argue that it is possible to identify a clear fundamental – and therefore risk-based factor structure.

We treat momentum as a fundamental phenomenon, driven by news-flow related changes to stock level, and expected returns of fundamentalist investors. We model two investor groups – short term and long term – sharing similar, but not always consistent views of fundamental value. The occasional, but sometimes significant, variance in their respective implied Sharpe ratios provides the trigger for a momentum phase, while their realignment signals its end. A third investor group, momentum traders, do not participate in the establishment, duration or ending of the momentum phase. This group are passive bystanders who, nonetheless, would benefit from knowledge of the signals generated.

Results generated from this analysis on the S&P500 stock universe from 2004 to 2019 confirm the outperforming nature of a momentum strategy based on this signal with high hit rates and annualised returns at the stock level and clear signs of convexity in the returns, consistent with the requirements of any ARP-based momentum portfolio. As a means of evaluating risk at the individual stock level and as a framework for systematic stock selection of momentum-based factor portfolios, the treatment of momentum as a stock level risk factor suggests that an implied Sharpe-ratio-based signal adds value to the decision-making processes of both traders and investors.

17 Stefano Cavaglia, Louis Scott, Kenneth Blay and Vincent de Martel

This chapter examines the use of commodity factors in a portfolio context, aimed at customising a portfolio for specific clients. A critical question is whether the inclusion of a factor overlay, such as momentum, carry or value, or a combination of these factors, can improve outcomes for aspiring retirees, compared to the simple approach of investing in a portfolio of equities and/or bonds. Using a bootstrap simulation approach, allowing for alternative scenarios for the performance of the assets and factors, it is found that the incorporation of a single factor overlay enhances the likelihood of an aspiring retiree achieving her goal. A combination of factors in the overlay provides additional diversification benefits for the more risk-averse retiree. It is argued that, due to time diversification over the business cycle, factor overlays enhance accumulated wealth net of costs

Contributors

Andrew Grant is a Sydney University academic whose areas of expertise are behavioural finance, individual investor decision making and betting markets, focusing on preference and belief-based asset allocation and asset-pricing decisions. He has also been engaged with industry, with studies of alternative finance and marketplace lending in the Asia-Pacific. He has appeared as a guest commentator in print, radio and on television, discussing issues such as gambling market and banking regulation, personal savings and asset allocation model evaluation.

Stephen Satchell is a fellow of Trinity College; he has consulted to a large number of financial institutions and published many papers. He believes that financial research should go beyond the normal offerings and he enjoys collaborating with scholars of other disciplines to advance his understanding of markets. He was the editor of Journal of Asset Management and is the co-founder of Quantess, an all women quant group as well as Head of Credit Research at Imagine, a fintech start-up offering imaginative mortgage products.

Nick Baltas, PhD, is a managing director and head of R&D of the Systematic Trading Strategies Group at Goldman Sachs. He is also a visiting researcher at Imperial College Business School and the executive editor of the Journal of Systematic Investing. In the past, he has held several positions both in the finance industry and in academia. His academic research has received numerous awards and has been published in peer-reviewed finance journals and books.

Ron Bird recently retired from both the University of Technology Sydney and Waikato University. His research interests have most recently concentrated on the mysteries of the funds management and retirement income industries. He has published widely in journals such as Management Science, the Journal of Portfolio Management and the Australian Journal of Management.

Kenneth Blay is Head of Thought Leadership for the Invesco Investment Solutions team. Prior to joining Invesco, Mr. Blay was the advisory research manager in the portfolio and risk research group at State Street Associates. Previously, he served as director of research at 1st Global. Mr. Blay earned a BBA in finance from the University of Texas at San Antonio. He has co-authored and published various works on asset allocation with Nobel Laureate Harry Markowitz.

Stefano Cavaglia holds a PhD in Finance from the Chicago Booth School of Business. Throughout his career, he has applied state-of-the-art finance theory. His early work led to the restructuring of the UBS global equity platform and the founding of a $1.3b hedge fund he managed at UBS O'Connor, Chicago. Presently, he is co-developing, with Invesco, novel multi-asset, factor-based completion portfolios. His research is widely cited in the CFA community and financial press.

Alessio de Longis is a Senior Portfolio Manager for the Invesco Investment Solutions division. He leads the group's tactical asset allocation, including factor and currency strategies. He joined Invesco from OppenheimerFunds, where he was team leader and senior portfolio manager of the multi-asset team. He earned an MSc in financial economics and econometrics from University of Essex, and MA/BA in economics from the University of Rome. He is a CFA Charterholder.

Vincent de Martel, CFA, is Senior Solutions Strategist with the Invesco Investment Solutions team. Prior to joining Invesco, Mr. de Martel was head of product strategy for BlackRock's multi-asset risk parity/factor suite. Previously, he served as head of European LDI strategy at Barclays Global Investors. He earned an MA degree in accounting and financial economics from the University of Essex, as well as an MBA from EDHEC with a concentration in market finance.

Dan diBartolomeo is President and founder of Northfield Information Services, Inc. Based in Boston since 1986, Northfield develops quantitative models of financial markets. He sits on boards of numerous industry organizations include IAQF and CQA, and is past president of the Boston Economic Club. His publication record includes 35 books, book chapters and research journal articles. In January of 2018, he became co-editor of the Journal of Asset Management. Dan spent numerous years as a Visiting Professor at Brunel University. In 2010 he was given the ‘Tech 40’ award by Institutional Investor magazine in recognition of his role in the discovery of the Madoff hedge fund fraud. He has also been admitted as an expert witness in litigation matters regarding investment management practices and derivatives in both US federal and state courts.

Kevin Gao is an experienced Business Intelligence Consultant with a demonstrated history of working in the information technology and services industry. He has analytical skills, and skills in quantitative research, data mining, banking, accounting and teaching. He has a PhD focused in Applied Finance, and degrees in mathematics and computer science.

Yang Gao joined Huazhong University of Science and Technology as assistant professor in June 2019 after obtaining his PhD from the University of Sydney. His research interests lie in the area of empirical asset pricing including trading strategies and market anomalies. He has been invited to present his work in multiple conferences including the American Finance Association annual meeting and the Goldman Sachs Asia Conference. Yang is also interested in corporate finance related studies with a focus on corporate governance.

Shivam Ghosh completed his PhD in Theoretical Physics at Cornell University in 2014 after obtaining a BA and MA in Natural Sciences (Physics) at the Cavendish Laboratory, University of Cambridge, UK. Shivam is now a Director Research at JPMorgan, London specialising in building tradable systematic risk premia strategies in credit. Previously, he was a fixed income portfolio manager at the Chief Investment Office, JPMorgan in New York. Shivam has published papers in top peer-reviewed journals including Physical Review B and Physical Review Letters.

Mo Haghbin, CFA, CAIA, serves as the Chief Operating Officer for the Invesco Investment Solutions team, which develops and manages customized multi-asset investment strategies. Mo Haghbin joined Invesco in 2019 when the firm combined with OppenheimerFunds, where he served as senior vice president and head of product for the Beta Solutions business. At OppenheimerFunds, he led the firm's entrance into factor investing, having launched a suite of equity factor strategies including one of the industry's first dynamic multi-factor ETFs.

Lei Ji is a senior quantitative analyst at Bloomberg. Before joining Bloomberg in 2015, Lei was a global quantitative portfolio manager at Tudor Investments. Prior to that, he was a founding member of the quantitative hedge fund GSB Podium Advisors. Before that, Lei served as a systematic proprietary trader at Bank of America, Merrill Lynch and Lehman Brothers. Lei received a PhD in Financial Economics from the University of Pennsylvania.

Robert Kosowski is Professor of Finance at Imperial College Business School and Head of Quantitative Research at Unigestion. He is also a research fellow at the Centre for Economic Policy Research (CEPR) and an associate member of the Oxford-Man Institute of Quantitative Finance at Oxford University. Robert is on the editorial board of the Journal of Alternative Investments and the Journal of Systematic Investing. Robert holds a BA (First Class Honours) and MA in Economics from Trinity College, Cambridge University (UK), and a MSc in Economics and PhD from the London School of Economics (UK).

Oh Kang Kwon is a senior lecturer in the Discipline of Finance at The University of Sydney. He holds a PhD in pure mathematics from MIT and a PhD in quantitative finance from UTS. He has held academic positions at various universities in Australia, and has also worked in the role of senior quantitative analyst with all four major banks in Australia. His research interests include quantitative finance, computational finance and derivative pricing. He has recently been working in the areas of corporate finance and strategy analysis

Muhammad Masood is a researcher in the iShares Product Development team at BlackRock. He is involved in the research and design of smart beta, ESG, thematic and fixed income ETF products. His work on smart beta include single- and multi-factor strategies that are constructed in a systematic manner. Muhammad holds a master's degree in Finance and Economics from the Warwick Business School and an undergraduate degree from the American University of Paris.

Jose Menchero serves as Head of Portfolio Analytics Research at Bloomberg. Prior to joining Bloomberg, he was Head of Equity Factor Model Research at MSCI Barra. Jose has roughly 40 finance publications in leading practitioner journals. Before entering finance, Jose was Professor of Physics at the University of Rio de Janeiro, Brazil. Jose holds a PhD in theoretical physics from the University of California at Berkeley and is a CFA Charterholder

Byoung-Kyu Min's primary research interests are in the areas of asset pricing and investments. His research aims to understand determinants of cross-sectional differences in stock returns, predictability of stock returns and its implication for trading strategies, and how financial markets and the macroeconomy (such as business cycles) are related and their implications for asset pricing. He proposes multiple explanations for financial market anomalies, including risk-based explanation, lottery preference and sentiment.

Vadim Moroz holds a PhD in Applied Mathematics from Northwestern University. He was co-founder and co-PM for the $1.3b UBS O'Connor Global Equity Long/Short strategy. Subsequently he managed high-frequency trading strategies at Tudor Capital, Citadel and JP Morgan. He is presently working as a Data Scientist supporting investment management functions. He was a co-recipient of the INQUIRE Europe prize for best research in 2003.

Anders Petterson began his career at JP Morgan and went on to set up ArtTactic in 2001. ArtTactic has become one of the leading art market research companies and a pioneer in using crowd-sourcing techniques for gathering and processing intelligence on the art market. Anders Pettersson is a regular lecturer on the art market for universities and businesses. He is also a founding board member of Professional Advisors to the International Art Market (www.paiam.org)

Christopher Polk is Department Head and Professor of Finance at the London School of Economics. He has published extensively in leading academic journals, receiving numerous professional awards including paper of the year twice at the Journal of Financial Economics. Polk has advised asset managers, central banks and regulators on topics related to his research. He holds a PhD in finance from the University of Chicago where he studied under 2013 Nobel Laureate Eugene Fama.

Katharina Schwaiger, PhD, is an investment researcher within the Factor-based Strategies Group at BlackRock. Her research responsibilities include macro and style factor research. Prior to joining BlackRock she worked as a Financial Engineer in the City of London, as a Quantitative Researcher at a London-based hedge fund and as a Lecturer at the London School of Economics. She received a BSc in Financial Mathematics and a PhD in Operational Research from Brunel University.

Louis Scott is the founder of Kiema Advisors, a research consultancy. He is currently working with Dr Cavaglia on research projects for Invesco. He was previously Head of FactSet's Risk and Quantitative Research. Louis has managed a global equities hedge fund at Old Mutual Asset Managers, US equities at Citigroup and currencies at PanAgora – each at over 1 billion in AUM. He is a member of the London Quant Group's management committee.

Nandini Srivastava completed her PhD at the Faculty of Economics, University of Cambridge in 2013 is now a Director at JPMorgan Asset Management, London, UK. She focuses on asset allocation research where her role involves quantitatively analysing the impact of macroeconomic and financial factors across a range of asset classes and developing strategies based on these for portfolio allocation. Nandini has been a referee at a number of academic journals and continues to publish in academic forums.

Christopher (Chris) Tinker is a founding partner of Libra Investment Services, an FCA regulated independent equity market research company. With more than 30-years experience in the Financial Services industry, he began his career as an Equity markets economist in the City of London before moving onto research roles in fixed income, credit, currency and money markets and as an international equity strategist in London and Hong Kong. He has a BA in Economics from Manchester University.

Danny Yeung is a lecturer at the UTS Business School. Danny's research interests concentrate in the area of investment, with particular emphasis on the impact of ambiguity and emotions on investor decision making, mutual fund performance and the impact of institutional investors in capital markets. He has published widely in international peer-reviewed journals including the Journal of Banking and Finance, the Pacific-Basin Finance Journal and the International Review of Financial Analysis.

Oliver J. Williams began his career at Boston Consulting Group, then worked in derivatives at JPMorgan and Credit Suisse before joining Markham Rae as partner, portfolio manager and lead architect/developer in systematic trading. He holds an MA in computer science and management studies, MPhil and PhD in financial economics from Cambridge University and MSc in mathematics from Birkbeck. He has co-authored several articles in investment management and is co-inventor of a software patent in network analytics

William E. (Bill) Zieff is Director at Northfield Information Services and was Managing Director, Chief Investment Officer and Portfolio Manager of Global Structured Products Group at Evergreen Investment Management Company. Previously, Bill was the Managing Director and Co-Chief Investment Officer of the Global Asset Allocation Group at Putnam Investments. He also served as a Director of Asset Allocation at Grantham, Mayo, Van Otterloo. Bill obtained a BSc in Economics and Mathematics from Brown University and an MBA from the Harvard Business School

Introduction

There are numerous definitions of momentum, which as an investment strategy is likely to be of great antiquity. Some authors credit Richard Driehaus with popularising the strategy, indeed he was described in the American Association of Individual Investors as the father of momentum investing. The concept of relative strength investing motivated Jegadeesh and Titman (1993) to analyse the momentum effect. Their basic finding showed that ‘winners’ (the top decile of performers) over the past three to twelve months continue to outperform ‘losers’ (the bottom decile) over the subsequent three to twelve months.

This return persistence phenomenon was initially documented through the following process. First, stocks are sorted each month into deciles based on performance over the past J months (the ‘formation’ or ‘ranking’ period). We might typically think of ‘Portfolio 10’ being that of the winning stocks, ‘Portfolio 9’ containing the second-best decile of performers, and so on down to ‘Portfolio 1’ containing the worst performing decile of stocks, ‘losers’.1

The momentum strategy involves holding ranked portfolios of the chosen stocks for the next K months (the ‘holding’ period). The version advocated by Jegadeesh and Titman involves taking a long position in the winners and shorting the losing stocks. Documented profits for an overlapping version of the strategy were estimated to be about 12% per year when and range between 6 and 12 months. In the opinion of Jegadeesh and Titman (2011, p. 494) the sustained and consistent magnitude of returns to a momentum strategy are too large to be explained by risk factors.

However, as this book will show, there are many different definitions of momentum, and we are attempting to understand them all within a common framework. Focusing on the above definition of momentum, it is clear that we can vary both the formation period and the holding period. There may also be an intervening period between of length between the two. Such a strategy could be defined by the triplet (), where the usual convention is for periods of , and months. However, any unit of time could be used. Further immediate generalisations could move from deciles to -tiles. One standard convention is that the portfolio is self-financing. This means that if you were to hold $100 million in winning stocks, you would short $100 million in losers.

I.1. HOW HAS MOMENTUM PERFORMED?

In Figure I.1, we present returns of value-weighted portfolios of US stocks using data from Ken French's website.2 The figure plots returns for two overlapping time periods, (i) the full sample period from January 1927 until November 2019, and (ii) the 30 year period from December 1989 until November 2019. The latter period coincides approximately with the timeframe since early versions of Jegadeesh and Titman's work became available.

To construct portfolios based on prior returns, stocks are sorted based on prior returns from month −12 to month −2. A month's return is withheld, and returns are reported on a monthly basis, which are then annualised and averaged. Using the above notation, we consider this an () strategy. We note that as we move from Losers (Portfolio 1) to Winners (Portfolio 10), the returns are virtually monotonically increasing in the ranking decile. The difference between the full sample period (solid shading in Figure I.1) and the 30-year period from the end of 1989 (striped columns in Figure I.1) appears to mainly be driven by the weakened performance of winners. Annual return to winners have declined by around 2.5% per annum (from approximately 18 to 15% pa), while losers have averaged approximately 4% per annum over both sample periods. Interestingly, there has been substantial ‘flattening’ of the portfolio returns in the remaining deciles. In the recent 30-year period, Portfolios 2 through 9 have averaged between 8 and 12% per annum.

FIGURE I.1 Annualised monthly returns for value-weighted portfolios formed based on prior returns (11,1,1). The figure reports the average returns to portfolios sorted on prior returns (where portfolio 1 is the portfolio of worst performing stocks, and portfolio 10 is the portfolio of best-performing stocks) over the period from January 1927 to November 2019 (solid bars) and December 1989 to November 2019 (striped shading).

TABLE I.1 Moments and percentiles of momentum portfolio returns over the period January 1927 to November 2019. Panel A reports average, standard deviations, skewness and the ratio of average returns to standard deviation (Ave/Std. Dev) for each of the 10 portfolios plus the momentum ‘Winner – Loser’ portfolio. Panel B reports monthly portfolio return percentiles for each of the portfolios.

Portfolio

Panel A: Moments (Annualised)

Average

  3.87

  8.33

  9.17

 10.57

 10.75

 11.49

 12.17

 13.38

 14.39

 17.94