by Dimitris Bertsimas, Allison K. O'Hair and William R. Pulleyblank
Dynamic Ideas, Belmont, Massachusetts, 2016.
The Analytics Edge provides a unified, insightful, modern and entertaining treatment of analytics. The book covers the science of using data to build models, improve decisions, and ultimately add value to institutions and individuals.
The philosophical underpinnings of the book are that real world problems are usually complex and often
will defined; they do not come with labels, meaning that they are not necessarily regression problems
or optimization problems. The only objective reality is data, which itself may be incomplete and of
questionable quality, and the role of models is to facilitate the solution of real world problems. Problems and data play a leading role in this book, while models play an essential but supporting role. This is in contrast with the vast majority of books and classes available today, in which methods play the leading role.
Distinguishing Characteristics of the Book
Most chapters in the book start with a real world problem (typically from our experience) and a data
set, and then use a variety of methods to address the problem. The book is organized into several
application areas in which analytics has had a significant impact. It is structured in seven parts:
- In Part I (Chapters 1-3), “Humans and Machines,” we discuss the competitive edge of models and machines versus humans in a variety of sometimes unexpected areas: predicting the quality of wine, assessing quality in healthcare, forecasting supreme court decisions, beating the world champion in chess, and beating the best human players in the game show Jeopardy!.
- In Part II (Chapters 4-6), “Sports and Games,” we present how analytics can be used in a variety of sport and game decisions. Specifically, we present how to evaluate championship players in both professional baseball and basketball, how to optimize the structure of the National Hockey League and how to provide an edge in blackjack.
- In Part III (Chapters 7-10), “Healthcare,” we show the impact that analytics has had in healthcare. We start with the Framingham Heart Study, which is to a large extent responsible for what we know today about heart disease. Then we demonstrate how analytics can be used to predict future healthcare costs, predict heart attacks, reduce mortality in infants and design an efficient and fair national policy for kidney allocations.
- In Part IV (Chapters 11-13), “The Internet,” we show how analytics is influencing the internet. We describe the heart of Google's search engine, how to optimize revenue from online advertising and how a variety of internet companies provide online recommendations,
- In Part V (Chapters 14-16), “Combating Crime,” we show how analytics can be used to combat crime. We describe how analytics can detect medicaid fraud, how to predict where crime will occur and how social networks can help reduce gang crime.
- In Part VI (Chapters 17-18), “Management of Operations,” we show how analytics provide an
edge in operations management. We discuss revenue management in airlines and casinos, and how to improve emergency room operations of a major hospital.
- In Part VII (Chapters 19-20), “Finance,” we show how analytics has had a substantial impact in finance and especially in the areas of asset management and the pricing of options.
- In Part VIII (Chapters 21-22), “Methods and Exercises,” we briefly discuss some of the analytics methods that are used in the book, and provide exercises to help the readers put analytics into
Using the Book in Course Design
This book can be used to teach several different types of courses to a variety of students. We have
used this book to teach a one semester class at the Massachusetts Institute of Technology (MIT) Sloan
School of Management for an audience that included MBA, undergraduate and graduate students. We
also taught a shorter class for Executive MBA students focusing on Chapters 1, 3, 4, 7, 8, 9, 11, 13, 14
and 19. A pre-requisite of these courses was a basic class on regression, probability and optimization.
However, we have also created a Massive Open Online Course (MOOC) through edX (15.071x), based
on this book that teaches the material with no assumed background knowledge.
Resources Available for all Readers
- Datasets for some chapters and the exercises.
- R Instruction Manual and R Script files.
You may click the button to initiate a file download.
Resources Available for Qualified Instructors
The following resources are available for instructors only. To obtain any of the resources below fill the Instructor Resource Form . If you have already filled the form and you have a password please click on this link .
About the Authors
Dimitris Bertsimas is the Boeing Professor of Operations Research and the co-director of the Operations Research Center at the Massachusetts Institute of Technology. He is a member of the National Academy of Engineering, recipient of numerous research awards and co-founder of six Analytics companies.
Allison K. O'Hair is a Lecturer in Operations, Information and Technology at the Stanford Graduate School of Business. She has a Ph.D. in Operations Research from MIT, and is a former Lecturer at the MIT Sloan School of Management.
William R. Pulleyblank is Professor of Operations Research at the United States Military Academy, West Point. Previously he was Vice President of Business Optimization in IBM and before that led the BlueGene supercomputing project in IBM Research. He is a member of the National Academy of
Engineering and has served on a broad range of business and government advisory boards.