George E P Box - An Accidental Statistician

Copyright 2013 by John Wiley & Sons, Inc. All rights reserved

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) 750-4470, 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 http://www.wiley.com/go/permission.

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 also publishes its books in a variety of electronic formats. Some content that appears in print may not be available in electronic formats. For more information about Wiley products, visit our web site at www.wiley.com.

Library of Congress Cataloging-in-Publication Data:

Box, George E. P.

An accidental statistician : the life and memories of George E.P. Box / George E. P. Box.

pages cm

Includes index.

ISBN 978-1-118-40088-3 (cloth)

1. Box, George E. P. 2. Statisticians\endash United States\endash Biography. I. Title.

QA276.157.B69A3 2013

519.5092dc23

[B]

2012040251

This book is dedicated to my students, with whom

it was my privilege to work, and who became my friends.

Virginia Woolf wrote about a character with a mind that kept throwing up from its depths, scenes, and names, and sayings, and memories and ideas, like a fountain spurting over. George Box is the embodiment of that active mind. Dinner with George is a spurting of stories, poems, songs, and anecdotes about his work and his friends. An Accidental Statistician jumps you into that fountain of ideas. It is great fun even if books about statistics and science are normally absent from your reading list.

No doubt many readers think, as I once did, that the subject is difficult and dull. Here we have a charming and colorful storyteller who quotes Yogi Berra in a discussion on the analysis of variance; has Murphy, of Murphy's Law fame, ringing the alarm when there is an opportunity to make things better; and explains an experiment with critical variables named banging and gooeyness. There are stories about composite designs, time-series forecasting, Evolutionary Operation, intervention analysis, and so on, but they are not mathematical, and most include personal anecdotes about people who were involved in their invention and original application. You learn about statistics and science and, simultaneously, meet a literal Who's Who of statisticians and scientists, and the Queen of England as well.

I met George Box in 1968 at the long-running hit show that he called The Monday Night Beer Session, an informal discussion group that met in the basement of his house. I was taking Bill Hunter's course in nonlinear model building. Bill suggested that I should go and talk about some research we were doing. The idea of discussing a modeling problem with the renowned Professor Box was unsettling. Bill said it would be good because George liked engineers. Bill and several of the Monday Nighters were chemical engineers, and George's early partnership with Olaf Hougen, then Chair of Chemical Engineering at Wisconsin, was a creative force in the early days of the newly formed Statistics Department. I tightened my belt and dropped in one night, sitting in the back and wondering whether I dared take a beer (Fauerbach brand, an appropriate choice for doing statistics because no two cases were alike). I attended a great many sessions over almost 30 years, during which hundreds of Monday Nighters got to watch George execute an exquisite interplay of questions, quick tutorials, practical suggestions, and encouragement for anyone who had a problem and wanted to use statistics. No problem was too small, and no problem was too difficult. The output from George was always helpful and friendly advice, never discouragement. Week after week we observed the cycle of discovery and iterative experimentation. We saw real examples that, All models are wrong, but some are useful. We saw how statistics is a catalyst for scientific method, and how scientific problems catalyze ideas for doing statistics. What a treat.

My business is water quality engineering. One night I wanted to discuss a problem that involved a measurement called the Biochemical Oxygen Demand (BOD). George asked whether it would be all right for him to explain this BOD test. He gave as good an explanation as I ever heard. I asked how he happened to know that and learned that at age 16, he took a job as an assistant chemist in a sewage treatment plant. One year before I was born, in 1939, at age 19, he published a paper about oxygen demand in the activated sludge wastewater treatment process, which at the time was new and poorly understood. George's paper can stand with papers on the subject written by some famous Wisconsin engineers who worked on the problem at about the same time. In the 1990s, 55 years later, George and I worked on forecasting the dynamics of activated sludge process performance using multivariate nonstationary time series. Imagine that from a world-famous statistician who was one of the earliest researchers on this widely used wastewater treatment process.

George and I have one bit of unfinished piece of business. A 20-foot-tall civil war soldier guards the stone arch entrance to Fort Randall Park, which is next to the engineering building and the statistics building at the University of WisconsinMadison. George had thought for some time that the soldier should have a medal. In 1993, we found a suitable brass medal in a sidewalk stall near Hyde Park, London, but our plan to hang it around the soldier's neck was never fulfilled. I now believe that the medal should stay with George as an award of merit for memoir writing. He deserves it. (And we two old friends do not have to climb the soldier.)

Last night, May 10, 2012, my wife and I had dinner with George and Claire at their house in Shorewood Hills. He said, The memoir is finished. I asked, What's your next project. It's hard to picture you not doing some writing every day. He answered, I 'm thinking of a paper about Fisher's idea on multiplicative effects in experiments.