Much has changed since we published the first edition of Finding Alphas, in 2015. The premise of that edition – that we considered these techniques “the future of trading” – is more true today than it ever was. In the intervening four years, we at WorldQuant have seen remarkable growth in our development of predictive algorithms for quantitative trading – we call them “alphas” – powered by an ever-rising volume and variety of available data, an explosion in computer hardware and software, and increasingly sophisticated techniques that allow us to create and deploy a higher volume and quality of alphas. Today, at WorldQuant, we have produced over 20 million alphas, a number that continues to grow exponentially as we hunt for ever-weaker predictive signals.

Since 2015, we have steadily expanded our outreach to new, diverse talent, adding to the full-time researchers, portfolio managers, and technologists at our 28 offices around the world, as well as to our growing group of research consultants – now more than 2,000 strong. We found many of WorldQuant's research consultants through our Virtual Research Center (VRC) and its global quantitative finance competitions, such as the WorldQuant Challenge, Women Who Quant, and the International Quant Championship.

Participants who enter our competitions seek to create high-quality alphas using our online portal, WebSim, which provides educational, research, and backtesting tools, including some of the same datasets that WorldQuant researchers use. More broadly, the VRC enables individuals to conduct research and seek to build high-quality algorithms that may be used in WorldQuant's systematic financial strategies. Research consultants have flexibility in their hours and work location, are compensated based on their activity and productivity, are eligible for additional compensation based on their algorithms' performance, and may ultimately be considered for full-time positions.

This book, with contributions from 47 current and former WorldQuant staffers, summarizes much of what we have learned about the art and science of alpha development. This edition is not just a new cover slapped on old content. Individual chapters have been extensively rethought and revised. This edition has nine chapters that didn't exist in 2015 – on exchange-traded funds, index alphas, intraday data, and event-driven investing, among other subjects – and extensive additions in most of the rest. Topics like machine learning and automated search have become much more important. But while we've gone deep, we've worked hard to make the material more accessible and useful.

Yet we're only beginning to plumb the possibilities of alphas – and to explore the universe of predictive signals. The years ahead will be full of new challenges, new data, and new techniques. This exponential world forces us to accept that what's here today won't necessarily be here tomorrow. In The UnRules: Man, Machines and the Quest to Master Markets, a book I published in 2018, I wrote that I have become convinced that the Age of Prediction is upon us. The more alphas you have, the better you can describe reality and the more predictive you can be. But change is a constant, and the task is never done.

This book is a study of the process of finding alphas. The material is presented as a collection of essays, providing diverse viewpoints from successful quants on the front lines of quantitative trading.

A wide variety of topics is covered, ranging from theories about the existence of alphas, to the more concrete and technical aspects of alpha creation.

Part I presents a general introduction to alpha creation and is followed by a brief account of the alpha life cycle and insights on cutting losses.

Part II focuses more on the technical side of alpha design, such as the dos and don’ts of information research, key steps to developing an alpha, and the evaluation and improvement of quality alphas. The key technical aspects discussed in this section are turnover, backtesting, fundamental analysis, equity price volume, statistical arbitrage, overfitting, and alpha diversity.

Part III explores ad hoc topics in alpha design, including alpha design for various asset classes like futures and currencies, the development of momentum alphas, and the effect of news and social media on stock returns.

In Part IV, we introduce you to WebSim, a web-based alpha development tool. We invite all quant enthusiasts to utilize this free tool to learn about alpha backtesting (also known as alpha simulation) and ultimately to create their own alphas.

Finally, in Part V, we present an inspirational essay for all quants who are ready to explore the world of quantitative trading.

In these pages, we present a collection of chapters on the algorithmic-based process of developing alphas. The authors of these chapters are WorldQuant's founder, directors, managers, portfolio managers, and quantitative researchers. This book has two key objectives: to present as many state-of-the-art viewpoints as possible on defining an alpha, and the techniques involved in finding and testing alphas. At WorldQuant, we believe that no viewpoint is the best and only answer, and that a variety of approaches is always superior to a single one.

This edition of Finding Alphas began to take shape in 2017, after Michael Peltz joined WorldQuant as global head of content. That year, he and then-regional research director Rebecca Lehman met with Igor Tulchinsky – WorldQuant's founder, chairman, and CEO – in his Connecticut office to outline their plan to overhaul the book. Finding Alphas had originally been written in 2014; Lehman and Peltz wanted to revise (and in some cases merge) existing chapters and add new ones. The two were instrumental in driving a reconceptualization of the first edition and managing a complex editorial process. We particularly want to thank the authors of the new chapters – Crispin Bui, Mark YikChun Chan, Chinh Dang, Glenn DeSouza, Anand Iyer, Rohit Kumar Jha, Michael Kozlov, Nitish Maini, Aditya Prakash, Prateek Srivastava, Dusan Timotity – as well as the authors of the chapters that were updated. Together they have created a book that should be tremendously useful to anyone interested in quantitative investing and developing alphas.

Every one of these chapters was copy edited by sharp-eyed Ruth Hamel, who juggled the myriad questions and challenges every chapter presented. Then each chapter was carefully vetted by WorldQuant's legal team under Jeffrey Blomberg, whose patience and always sensible suggestions made a major contribution. And, once again, we thank Wendy Goldman Rohm, our literary agent, who played a major role in getting the first edition of Finding Alphas off the ground.

Last, we need to acknowledge with gratitude the support and faith of every colleague at WorldQuant. It takes a team. Thank you all.

DISCLAIMER

The contents of this book are intended for informational and educational purposes only and, as such, are not intended to be nor should be construed in any manner to be investment advice. The views expressed are those of the various contributors and do not necessarily reflect the view or opinion of WorldQuant or the WorldQuant Virtual Research Center.

At the time of writing, the WebSim information contained in this book is consistent with the WebSim website. Because the website is subject to change, in cases where there are inconsistencies between this book and the website the terms of the WebSim website will govern the most updated and current processes of WebSim. For the most up-to-date version of WebSim and the terms applicable to its use, please go to https://worldquantvrc.com or its successor site.

Registration at WebSim's official website is required to obtain the full functionality of the platform and to have access to the WebSim support team. Successful alphas submitted by research consultants may, in certain cases, be considered for inclusion in actual quant trading investment strategies managed by WorldQuant.

WEBSIM RESEARCH CONSULTANTS

WorldQuant has established a Research Consultant program for qualified individuals to work with our web-based simulation platform, WebSim. This program gives consultants the flexibility to create alphas in their own physical and intellectual environment. This is a particularly ideal pursuit for individuals who are undertaking a college education, as well as those who are ambitious and highly interested in breaking into the financial industry.

Qualified candidates are those highly quantitative individuals who typically come from science, technology, engineering, or mathematics (STEM) programs. However, majors and expertise vary and may include statistics, financial engineering, math, computer science, finance, physics, or other STEM programs.

You can find more details on WebSim in Part IV of this book. More information on the Research Consultant program is available at WebSim's official website.

What is an alpha? Throughout this book, you'll read different descriptions or definitions of an alpha. Alpha, of course, is the first letter of the Greek alphabet – as in “the alpha and the omega,” the beginning and the end – and it lurks inside the word “alphabet.” Over the centuries, it has attached itself to a variety of scientific terms. The financial use of the word “alpha” goes back to 1968, when Michael Jensen, then a young PhD economics candidate at the University of Chicago, coined the phrase “Jensen's alpha” in a paper he published in The Journal of Finance. Jensen's alpha measured the risk-adjusted returns of a portfolio and determined whether it was performing better or worse than the expected market. Eventually, Jensen's alpha evolved into a measure of investment performance known simply as alpha, and it is most commonly used to describe returns that exceed the market or a benchmark index.

Since then, the term “alpha” has been widely adopted throughout the investing world, particularly by hedge funds, to refer to the unique “edge” that they claim can generate returns that beat the market. At WorldQuant, however, we use the term a little differently. We design and develop “alphas” – individual trading signals that seek to add value to a portfolio.

Fundamentally, an alpha is an idea about how the market works. There are an infinite number of ideas or hypotheses or rules that can be extrapolated, and the number of possibilities is constantly growing with the rapid increase in new data and market knowledge. Each of these ideas could be an alpha, but many are not. An alpha is an automated predictive model that describes, or decodes, some market relation. We design alphas as algorithms, a combination of mathematical expressions, computer source code, and configuration parameters. An alpha contains rules for converting input data to positions or trades to be executed in the financial securities markets. We develop, test, and trade alphas in large numbers because even if markets are operating efficiently, something has to drive prices toward equilibrium, and that means opportunity should always exist. To use a common metaphor, an alpha is an attempt to capture a signal in an always noisy market.

DESIGNING ALPHAS BASED ON DATA

We design alphas based on data, which we are constantly seeking to augment and diversify. Securities prices generally change in response to some event; that event should be reflected in the data. If the data never changes, then there is no alpha. Changes in the data convey information. A change in information should in turn produce a change in the alpha. These changes may be expressed in a variety of alpha expressions. Table 1.1 shows a few simple examples.

Alpha design is really just the intelligent search for price information conveyed by possible changes in the data, whether you think of them as patterns, signals, or a code. The mathematical expression of an alpha should embody a hypothesis or a prediction. Again, just a few examples are shown in Table 1.2.

In the field of finance, an alpha is the measure of the excess return of an investment over a suitable benchmark, such as a market or an industry index. Within the quantitative investment management industry, and in this book, the term “alpha” refers to a model used to try to forecast the prices, or returns, of financial instruments relative to a benchmark. More precisely, an alpha is a function that takes, as input, data that is expected to be relevant to the prediction of future prices and outputs values corresponding to the forecasted prices of each instrument in its prediction universe, relative to a benchmark. An alpha can be expressed as an algorithm and implemented in a computer language such as C++, Python, or any number of alternative modern or classical programming languages.

Attempts to forecast markets predate the digital era and the arrival of computers on Wall Street. For example, in his 1688 treatise on economic philosophy, Confusion of Confusions, stock operator and writer Josseph Penso de la Vega described valuation principles for complex derivatives and techniques for speculating on the Amsterdam Stock Exchange. Two hundred years later, in a series of articles, Charles Dow (co-founder of Dow Jones & Co., which publishes The Wall Street Journal) codified some of the basic tenets of charting and technical analysis. His writings provide one of the first recorded instances of a systematic market forecasting technique, but investors had to wait until the 1980s for affordable computing power to arrive on the Wall Street scene and change the modeling paradigm: instead of pencil and paper, the main design tools and their hardware were set to become computers and digital data.