Cover

Contents

Cover

series

Title Page

Foreword

Preface

HOW THE BOOK IS ORGANIZED

Acknowledgments

Introduction: Why Risk Management Is Mostly Misunderstood

QUANTITATIVE RISK MANAGEMENT BEGINNINGS

QUANTITATIVE RISK MANAGEMENT SUCCESSES

QUANTITATIVE RISK MANAGEMENT FAILURES

WARREN BUFFETT’S RISK MANAGEMENT STRATEGY

DEFINING RISK MANAGEMENT

FAT TAILS, STATIONARITY, CORRELATION, AND OPTIMIZATION

MANAGING THE RISKS OF A RISK MANAGEMENT STRATEGY

THE RISK MANAGEMENT OPPORTUNITY SET

NOTES

Part One

Chapter 1: Exposed versus Experienced Risk Revisited

EXPOSURE HEDGE VERSUS DOLLAR HEDGE

HOW THE CREDIT CRISIS MOVED RISK MANAGEMENT TO THE FOREFRONT

RISKS BEYOND VOLATILITY

WHAT RISK MANAGEMENT SHOULD PROVIDE

CLARIFYING EXPECTATIONS OF RISK MANAGEMENT

AN EXAMPLE

NOTES

CHAPTER 2: Definitions of Tractable Risk

THE EFFECT OF UNCERTAINTY ON OBJECTIVES

IDENTIFYING AND MEASURING RISKS

FORECASTING AND HEDGING RISKS

PORTFOLIO VIEW VERSUS SECURITY-LEVEL VIEW

TOTAL RISK VIEW OF MULTI-ASSET-CLASS (MAC) PORTFOLIOS

STABILITY AND ACCURACY

NOTE

CHAPTER 3: Introduction to Asset Class Specifics

EQUITIES

FIXED INCOME

CONCLUSION

NOTES

CHAPTER 4: Commodities and Currencies

COMMODITIES

INTRODUCTION TO CURRENCY RISK

CONCLUSION

NOTES

CHAPTER 5: Options and Interest Rate Derivatives

SHORT HISTORY OF OPTION PRICING

VOLATILITY SMILE

IMPLIED VOLATILITY MODEL

BARONI-ADESI WHALEY (BAW) OPTION PRICING METHODOLOGY

OTHER OPTION PRICING METHODS

SWAPS, SWAPTIONS, FORWARDS, AND FUTURES

CONCLUSION

NOTES

CHAPTER 6: Measuring Asset Association and Dependence

THE SAMPLE COVARIANCE MATRIX

ESTIMATION ERROR MAXIMIZATION

MINIMIZING THE EXTREMES

THE COPULA, THE MOST COMPREHENSIVE DEPENDENT STRUCTURE MEASURE

THE MODEL COVARIANCE MATRIX

NOTES

CHAPTER 7: Risk Model Construction

MULTIFACTOR PRESPECIFIED RISK MODELS

PRINCIPAL COMPONENT (STATISTICAL) RISK MODELS

CUSTOMIZED HYBRID RISK MODELS

NOTES

Part Two

CHAPTER 8: Fixed Income Issues

VARIETY. ILLIQUIDITY. SIZE.

EMPIRICAL EVIDENCE

TEST PORTFOLIOS AND METHODOLOGY

TEST METRICS

COMPUTATIONAL EFFICIENCY

CONCLUSION

NOTES

CHAPTER 9: Interest Rate Risk

THE TERM STRUCTURE

TERM STRUCTURE DYNAMICS

FACTOR MODELS

STOCHASTIC DIFFERENTIAL EQUATIONS

INTEREST RATE RISK EXPOSURES

RISK FORECASTING

CONDITIONAL DURATION AND EXPECTED TAIL DURATION

CONCLUSION

NOTES

CHAPTER 10: Spread Risk

SPREAD BASICS

REDUCED FORM APPROACH

STRUCTURAL APPROACH

SPREAD EXPOSURE

SPREAD VOLATILITY

DERIVED SPREAD APPROACH

EURO-SOVEREIGN SPREADS

FACTOR MODEL APPROACH

CONCLUSION

NOTES

CHAPTER 11: Fixed Income Interest Rate Volatility, Idiosyncratic Risk, and Currency Risk

FIXED INCOME INTEREST RATE RISK

FIXED INCOME IDIOSYNCRATIC BOND RISK

FIXED INCOME CURRENCY RISK

CONCLUSION

NOTES

CHAPTER 12: Portfolio Risk Measures

COHERENT RISK MEASURES

COMMONLY USED RISK MEASURES

MARGINAL CONTRIBUTION

STRESS-TESTING

NOTES

Chapter 13: Risk for the Fundamental Investor

FUNDAMENTAL INVESTING VERSUS OTHER APPROACHES

TYPICAL RISK CONTROLS FOR FUNDAMENTAL INVESTORS

IMPLEMENTING RISK MANAGEMENT STRATEGIES INTO A FUNDAMENTAL PROCESS

OPTIMIZATION

CONCLUSION

Chapter 14: Portfolio Optimization

THE ENHANCED MVO MODEL

CONSTRAINTS AND OBJECTIVES IN EMVO

FURTHER IMPROVEMENTS TO THE ENHANCED MVO MODEL

FACTOR ALIGNMENT PROBLEMS

CONSTRAINT ATTRIBUTION

SPECIALLY STRUCTURED MVO MODELS

EXTREME TAIL LOSS OPTIMIZATION

INCORPORATING NONLINEAR INSTRUMENTS IN THE EMVO MODEL

ALGORITHMS FOR SOLVING MVO MODELS

HOW TO CHOOSE AN OPTIMIZER

NOTES

Part Three

Chapter 15: The SunGard APT Risk Management System

INTRODUCTION TO STATISTICAL FACTOR MODELS

APT FACTOR MODEL ESTIMATION—EQUITIES MODELS

SELECTION OF THE CORE UNIVERSE FOR FACTOR MODELING

CHOOSING THE NUMBER OF APT FACTORS

ESTIMATING THE RISK PROFILES IN AN APT FACTOR MODEL

APT MULTI-ASSET-CLASS FACTOR MODEL ESTIMATION

MODELING DERIVATIVES AND OTHER NONUNDERLYING SECURITIES

USER-DEFINED ASSETS WITHIN APT MODELS

CONCLUSION

NOTES

Chapter 16: Axioma Risk Models

BACKGROUND

RISK MODEL–BASED REPORTING

ROLE OF RISK MODELS IN INVESTMENT DECISIONS

AXIOMA VALUE AT A HIGH LEVEL

DAILY RISK MODELS, DELIVERED DAILY

MULTIPLE RISK MODELS

EMPIRICAL RESULTS

DETAILS OF AXIOMA INNOVATIONS

CONCLUSION

NOTES

Chapter 17: Distinguishing Risk Models

HISTORY

RISK MODEL DETAILS

RISK MODEL–BASED REPORTING

CONCLUSION

NOTES

Chapter 18: Northfield’s Integration of Risk Assessments across Multiple Asset Classes

A UNIFIED FRAMEWORK

INTEREST RATE RISK

CREDIT RISK

EQUITY FACTOR REPRESENTATION OF CORPORATE CREDIT RISK

DEFAULT CORRELATION

COMPLEX INSTRUMENTS AND DERIVATIVES

PRIVATE EQUITY

DIRECT REAL ESTATE AND GEOGRAPHICALLY LOCALIZED ASSETS

CONCLUDING EXAMPLE

CONCLUSION

REFERENCES

Chapter 19: R-Squared

WHY BUILD STOCK RISK MODELS?

GENERIC RISK MODELING

PRACTICAL RISK MODELING

STATISTICAL FACTOR MODELS

DEFINED FACTOR MODELS

ESTIMATE FACTORS OR ESTIMATE BETAS?

PRACTICAL CONSEQUENCES AT THE STOCK LEVEL

PRACTICAL CONSEQUENCES AT THE PORTFOLIO LEVEL

A SHORT DIGRESSION

HYBRID RISK MODELS

THE R-SQUARED SHORT-TERM HYBRID RISK MODEL FOR GLOBAL EQUITIES

SUMMARY

NOTE

Chapter 20: The Future of Risk Management and Analytics

THE INCREASING REGULATORY ENVIRONMENT

THE IMPACT OF REGULATIONS WITH TECHNOLOGY

THE FUTURE VIEW

NEW TYPES OF RISK MODELS

STRESS-TESTING YOUR WAY TO EVENT RISK PREPAREDNESS

Index

Cover design: John Wiley & Sons

Published by John Wiley & Sons, Inc., Hoboken, New Jersey.
Published simultaneously in Canada.

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, scanning, or otherwise, except as permitted under Section 107 or 108 of the 1976 United States Copyright Act, without either the prior written permission of the Publisher, or authorization through payment of the appropriate per-copy fee to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, (978) 750-8400, fax (978) 646-8600, or on the Web at www.copyright.com. Requests to the Publisher for permission should be addressed to the Permissions Department, John Wiley & Sons, Inc., 111 River Street, Hoboken, NJ 07030, (201) 748-6011, fax (201) 748-6008, or online at www.wiley.com/go/permissions.

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. No warranty may be created or extended by sales representatives or written sales materials. The advice and strategies contained herein may not be suitable for your situation. You should consult with a professional where appropriate. Neither the publisher nor author shall be liable for any loss of profit or any other commercial damages, including but not limited to special, incidental, consequential, or other damages.

For general information on our other products and services or for technical support, please contact our Customer Care Department within the United States at (800) 762-2974, outside the United States at (317) 572-3993 or fax (317) 572-4002.

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.

Library of Congress Cataloging-in-Publication Data:

Greiner, Steven P.
Investment risk and uncertainty : advanced risk awareness techniques for the intelligent investor / Steven P. Greiner.
pages cm. — (Wiley finance series)
Includes index.
ISBN 978-1-118-30018-3 (cloth); ISBN 978-1-118-41966-3 (ebk);
ISBN 978-1-118-42141-3 (ebk); ISBN 978-1-118-65373-9 (epdf)
1. Investments. 2. Risk. 3. Investment analysis. I. Title.
HG4521.G7234 2013
332.601–dc23
2012040847

Foreword

For many years, I have written and edited finance books as well as read the works of many others covering a wide range of topics. What is immediately recognizable about this book, however, is that it touches on many difficult and deep topics all under the same roof. Investing Risk and Uncertainty is able to do so by gathering the expertise of several highly qualified contributors and seamlessly organizing it all in one place for the reader.

Although the book does have some financial mathematics throughout, that is not the focus—the core message is about multi-asset class risk measurement, on-going monitoring, and mitigation. It is a formal practitioner’s guide to the topic of risk, from first principle’s definitions through to the latest mathematical construct of sophisticated risk measures calculations. To deliver that topic, the book starts with simple definitions of risk and in particular, introduces risk, its management, and its interpretation in historical context. This is an essential element often left out of books on the subject, leading a less informed reader up the learning curve much quicker than they might arrive naturally.

Since the Great Financial Crisis of 2008, there has been much discussion in the media about risk management failures. Likewise this on-going crisis has seen the promotion of risk managers to positions offering more authority over portfolio assets than they’ve ever had before. This has led to greater attention paid to the risks of portfolios and assets along with the development of new techniques for risk estimation. But the context of why that’s necessary is completely missed if one doesn’t understand the current useful measures of risk, understand their assumptions, their credible theories and how the current suite of risk estimation procedures came to be. The book’s main author Steven Greiner captures that sentiment when he says:

When financial failure happens, everybody’s life is touched because of the massive amount of money that is involved. As a result, every carbon-based life unit on the planet feels they have something valuable to say about it. Opinion and baseless facts construct the everyday perception of the situation since they are the dominate voices, and the realities are morphed far away from the truth.

Thus, this volume seeks first to offer a wider perspective on which to understand why risk management is necessary, beginning in its early days, and ending with explaining why we need more of it in the future.

The book’s main achievement is its ability to form a co-operative of competing risk vendors assessment of methodologies available for measuring risk. Nowhere else can one flow from SUNGARD-APT’s to R-Squared’s risk model(s) while covering Northfield’s, Axioma’s, and Barra’s along the way, in a single work and have their founders and experts in the risk community work together to form a complete volume on risk. Surely this feat alone must mean the crisis of the last decade and its systemic risks to the field and business must have been obscenely egregious compared to past financial crisis.

Finally, every investor needs global exposure these days and in many different asset classes. The largest companies in the world are no longer “attached” to their local economies. Globalization is not only being encouraged, it cannot be stopped. Technology has made the world too small and pensions, hedge funds and every other investor is thinking globally, if not traveling globally unlike ever before in history. Before recently, most firms measured or estimated risk in specific asset classes separately, then aggregated the risk to attempt full risk estimation across all asset types and portfolios. This was done first in countries and regions, and then combined to form their global risk assessment. In doing so, the risks were obviously over-estimated simply because diversification across assets was not ascertained. Recently though due to the aforementioned crisis, attention has been on forming better estimates of risks globally and across asset types. This involves a careful estimation of the covariance matrix, or more importantly, asset association of which covariance is just one type of measure. Additionally, the software and hardware technology is just beginning to allow these very large calculations to be performed in “almost” real-time and there is pressing need for it. This book addresses all of the major concerns with this kind of overall risk measuring process and on some level covers every topic related to it. I cannot think of a more comprehensive and concise book on the subject.

Within this volume, the authors set about writing the requirements for a multi-asset class global risk model and do so with a good mixture of theory and real world examples. Practitioners on every level should prepare themselves to be pleasantly surprised by the information found here and the unique, collaborative approach of its authors.

FRANK J. FABOZZI
January 2013

Preface

HOW THE BOOK IS ORGANIZED

This book is divided into three parts. The first part will discuss risk at a very high level, 30,000 feet if you reside in the United States and 10.000 meters if you live somewhere else. The beginning chapters will lay the groundwork giving the perspective on risk, outlining the hat you should wear when beginning to think about applying a risk management strategy. In this part we’ll describe risk exposures versus experienced risks, provide the working definitions of the risk measures (what exactly we are calculating at the portfolio level), and then delve qualitatively into various asset classes to bring out the fat, those verbal descriptions of what is risky about these assets, what is a calculably reliable measure of risk, and what isn’t. Thus the specifics of risk measuring for equities, the myriad fixed income types of securities, commodities, currencies, futures, and options will all be given a risk overview, definitions made, and the various ways of incorporating their pricing algorithms for risk forecasting explained. Interest rate derivatives will be reviewed as well.

The covariance matrix will be introduced, discussed, and dissected; and its comparison to using a copula will be explained and their respective trade-offs reviewed. Descriptions of multifactor risk models, which are the most common type of risk model, will be explained. Risk models created using principal component analysis will be introduced, and how one combines the two to produce hybrid risk models will be reviewed. Part One will have little or no mathematics in it, but the results of mathematical computations may be shown from time to time.

The second part will develop a more rigorous approach to defining risks. This part starts out with raising the issues associated with developing risk models for fixed income assets. A large part of this discussion will address each of the individual risks that a bond offers. We’ll address the risks that are covered in FactSet and what’s needed to be able to calculate them. We’ll discuss risk model construction, historical risk, and the ARCH and GARCH techniques, showing some examples of how these work as of today’s computational capabilities, their usage, and where and how they’ll probably be used in the future.

The mostly widely used and commonly reported risk measures will be discussed and formulas for their calculation will be shown. Then, and throughout, some computational modeling pitfalls will continuously be reviewed. A whole chapter is dedicated to how fundamental managers can use risk models, and will identify insights that risk models bring to the investment process specifically from the perspective of fundamental managers. In this chapter we’ll discuss what they can use the covariance matrix for and how it can be used to ascertain and shine a light on risks they’re not used to examining directly. This is where the “Warren Buffett is a covariance matrix” analogy will be (sort of) emphasized.

When it comes to risk forecasting, a very commonly used methodology involves mean-variance portfolio optimization, especially for equities. We’ll have a chapter on optimization, which will be reviewed in significant detail. Part Two will have lots of mathematics and statistics, but presumption on the audience will be avoided as much as possible, which we’ll strongly emphasize.

Part Three will review the risk products available in FactSet with working examples. This part will provide detailed overviews of the most prominent risk modeling capabilities within the industry. Vendors of renown such as Axioma, SunGard APT, and R-Squared will be featured prominently. For its historical significance, Barra will be discussed, as it was the first risk vendor, launched by Barr Rosenberg in the mid-1970s. Professor Rosenberg pioneered the use of quantitative methods for formulating risk metrics and investment strategies, and no discussion about equity risk would be complete without this review.

The book concludes with some modest forecasts of our own. Although most readers are aware that prognosticators and fortune-tellers have a large standard error between forecasts and realization, we will utilize some obvious trends combined with future computing speed (which never ceases to increase) to postulate where risk management might find itself in a few years.

A variety of authors and editors contributed to this encyclopedic detailing of risk models and management. It’s on shoulders I stand and it’s with their expertise that I offer this exposition for the benefit of an industry I love so much.

WATCH THE VIDEOS

Investment Risk and Uncertainty is accompanied by eight videos you can watch on your computer or mobile device. Three of the videos are FactSet or other risk vendor heads of research answering questions put to them on a conference risk panel. The remaining videos are how-to videos for using FactSet products.

When you see a box like this in your text, you will be directed to a video that relates to what you are reading. Go to www.wiley.com/go/greiner to see the full listing.

Acknowledgments

This book would not have been possible without the support from FactSet Research Systems and particularly from Chris Ellis, the senior head of product development, who in many ways is the original thought leader for risk on FactSet. Likewise, the chapter authors from FactSet’s risk vendors include Jason MacQueen, Laurence Wormald, Sebastian Ceria, and Daniel diBartolomeo, to whom we owe a hearty thanks for offering their expertise freely and also for their advice from time to time on technical matters and with clients. Sebastian graciously wrote the chapter on optimization, which was a special help, along with his colleague Kartik Sivaramakrishnan. Additionally, Melissa Brown and Bill Wynne from Axioma collaborated with Sebastian on writing a very user-friendly, easy-to-read chapter with good examples on using their products within FactSet.

I have great pleasure working at FactSet, which has some of the nicest people on the planet. These include the authors and editors Bill McCoy, David Mieczkowski, Richard Barrett, Christopher Carpentier, Mido Shammaa, Sameer Patel, Roberto Isch, Joe Importico, Viviana Vieli, Daniel Mathon, and Andrew Geer. Andrew in particular spent an inordinate amount of his time editing many chapters and offering help in formatting throughout the text. Katherine McCabe, Michelle Bova, and Mathew Ward also contributed, and it is Mathew’s magic that makes the product videos sprinkled throughout the book come to life. Without these people, this book would not have been written, let alone a hope or possibillity. Their contributions have been substantial.

I speak for all the authors collectively that we hope our clients use this book in such a way as to aid them in improving their investment process. Moreover, many of the issues we are involved with day to day are of such detail that the often cited 30-minute phone call with a client just doesn’t do them justice. For this reason I was inspired to write this book for their longer-term benefit, heavily assisted by the aforementioned contributors, who gladly joined my vision. It is hoped this volume will sit on clients’ desks as a quick and detailed reference for many years to come.

On a personal level, I dedicate this book to my daughter Vanessa. As of this writing she has started her first year in college, where I wish her more than luck and my sincerest well-wishing, even above what my heart cannot evoke in words.

Last, my wife Veronica continues to offer a steady hand, which I know I lean on more than I’d like to admit. Her love is constant, inveterate, and resolute. A thank you to her doesn’t convey properly all that she deserves.