Get the full ebook with Bonus Features for a Better Reading Experience on ebookluna.

com

Modern Business Analytics: Practical Data Science

for Decision-making - eBook PDF

https://ebookluna.com/download/modern-business-analytics-
practical-data-science-for-decision-making-ebook-pdf/

OR CLICK HERE

DOWLOAD NOW

Download more ebook instantly today at https://ebookluna.com

 Instant digital products (PDF, ePub, MOBI) ready for you
Download now and discover formats that fit your needs...

(eBook PDF) Business Analytics Data Analysis Decision

Making 6th

https://ebookluna.com/product/ebook-pdf-business-analytics-data-
analysis-decision-making-6th/

ebookluna.com

Business Analytics: Data Analysis & Decision Making 6th

Edition (eBook PDF)

https://ebookluna.com/product/business-analytics-data-analysis-
decision-making-6th-edition-ebook-pdf/

ebookluna.com

(eBook PDF) Business Analytics: Data Analysis & Decision

Making 7th Edition

https://ebookluna.com/product/ebook-pdf-business-analytics-data-
analysis-decision-making-7th-edition/

ebookluna.com

Business Analytics: Data Analysis & Decision Making

(MindTap Course List) 7th Edition Albright - eBook PDF

https://ebookluna.com/download/business-analytics-data-analysis-
decision-making-mindtap-course-list-ebook-pdf/

ebookluna.com
(eBook PDF) Analytics, Data Science, & Artificial
Intelligence: Systems for Decision Support 11th Edition

https://ebookluna.com/product/ebook-pdf-analytics-data-science-
artificial-intelligence-systems-for-decision-support-11th-edition/

ebookluna.com

Analytics, data science, & artificial intelligence :

systems for decision support Eleventh Edition Dursun Delen
- eBook PDF
https://ebookluna.com/download/analytics-data-science-artificial-
intelligence-systems-for-decision-support-ebook-pdf/

ebookluna.com

(eBook PDF) Accounting Business Reporting for Decision

Making 6th

https://ebookluna.com/product/ebook-pdf-accounting-business-reporting-
for-decision-making-6th/

ebookluna.com

(eBook PDF) Statistics for Business Decision Making

Analysis 2nd

https://ebookluna.com/product/ebook-pdf-statistics-for-business-
decision-making-analysis-2nd/

ebookluna.com

(eBook PDF) Spreadsheet Modeling & Decision Analysis: A

Practical Introduction to Business Analytics 8th Edition

https://ebookluna.com/product/ebook-pdf-spreadsheet-modeling-decision-
analysis-a-practical-introduction-to-business-analytics-8th-edition/

ebookluna.com
ISTUDY
 MODERN BUSINESS
ANALYTICS

ISTUDY
ISTUDY
 MODERN BUSINESS
ANALYTICS
Practical Data Science for Decision Making

Matt Taddy
Amazon, Inc.

Leslie Hendrix
University of South Carolina

Matthew C. Harding
University of California, Irvine

ISTUDY
 Final PDF to printer

MODERN BUSINESS ANALYTICS

Published by McGraw Hill LLC, 1325 Avenue of the Americas, New York, NY 10019. Copyright ©2023 by
McGraw Hill LLC. All rights reserved. Printed in the United States of America. No part of this publication may
be reproduced or distributed in any form or by any means, or stored in a database or retrieval system, without
the prior written consent of McGraw Hill LLC, including, but not limited to, in any network or other electronic
storage or transmission, or broadcast for distance learning.
Some ancillaries, including electronic and print components, may not be available to customers outside the
United States.
This book is printed on acid-free paper.
1 2 3 4 5 6 7 8 9 LKV 27 26 25 24 23 22
ISBN 978-1-266-10833-4
MHID 1-266-10833-5
Cover Image: MirageC/Getty Images

All credits appearing on page or at the end of the book are considered to be an extension of the copyright page.

The Internet addresses listed in the text were accurate at the time of publication. The inclusion of a website
does not indicate an endorsement by the authors or McGraw Hill Education, and McGraw Hill Education
does not guarantee the accuracy of the information presented at these sites.

mheducation.com/highered

ISTUDY tad08335_fm_ise.indd iv 01/28/22 07:31 PM

 ABOUT THE AUTHORS

Matt Taddy is the author of Business Data Science

(McGraw Hill, 2019). From 2008–2018 he was a professor of econo-
metrics and statistics at the University of Chicago Booth School of
Business, where he developed their Data Science curriculum. Prior
to and while at Chicago Booth, he has also worked in a variety of
industry positions including as a principal researcher at Microsoft and
a research fellow at eBay. He left Chicago in 2018 to join Amazon as
Courtesy of Matt Taddy a vice president.

Leslie Hendrix is a clinical associate professor in the

Darla Moore School of Business at the University of South Carolina.
She received her PhD in statistics in 2011 and a BS in mathematics
in 2005 from the University of South Carolina. She has received two
university-wide teaching awards for her work in teaching business ana-
lytics and statistics courses and is active in the research and teaching
communities for analytics. She was instrumental in founding the Moore
Courtesy of Leslie Hendrix School’s newly formed Data Lab and currently serves as the assistant
director.

Matthew C. Harding is a professor of economics

and statistics at the University of California, Irvine. He holds a PhD
from MIT and an M.Phil. from Oxford University. Dr. Harding conducts
research on econometrics, consumer finance, health policy, and energy
economics and has published widely in leading academic journals. He
is the founder of Ecometricx, LLC, a big data and machine learning
consulting company, and cofounder of FASTlab.global Institute, a
Courtesy of Matthew nonprofit focusing on education and evidence-based policies in the
C. Harding areas of fair access and sustainable technologies.

ISTUDY
 BRIEF CONTENTS

About the Authors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v

Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x
Guided Tour . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi
Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xvii

1 Regression. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

2 Uncertainty Quantification. . . . . . . . . . . . . . . . . . . . . . . . . . . 55

3 Regularization and Selection. . . . . . . . . . . . . . . . . . . . . . . . 100

4 Classification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151

5 Causal Inference with Experiments . . . . . . . . . . . . . . . . . . 175

6 Causal Inference with Controls. . . . . . . . . . . . . . . . . . . . . . 215

7 Trees and Forests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 259

8 Factor Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 282

9 Text as Data. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 316

10 Deep Learning. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 345

Appendix: R Primer. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 383

Bibliography 419
Glossary 424
Acronyms 433
Index 435

vi

ISTUDY
 CONTENTS

About the Authors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v

Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x
What Is This Book About? x

Guided Tour . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi
Practical Data Science for Decision Making xi
An Introductory Example xii
Machine Learning xiv
Computing with R xv

Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xvii
1 Regression . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Linear Regression 3
Residuals 15
Logistic Regression 21
Likelihood and Deviance 26
Time Series 30
Spatial Data 46

2 Uncertainty Quantification. . . . . . . . . . . . . . . . . . . . . . . . . . . 55
Frequentist Uncertainty 56
False Discovery Rate Control 67
The Bootstrap 72
More on Bootstrap Sampling 86
Bayesian Inference 91
3 Regularization and Selection. . . . . . . . . . . . . . . . . . . . . . . . 100
Out-of-Sample Performance 101
Building Candidate Models 108
Model Selection 130
Uncertainty Quantification for the Lasso 144

4 Classification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
Nearest Neighbors 152
Probability, Cost, and Classification 158
Classification via Regression 160
Multinomial Logistic Regression 163

ISTUDY
 viii    Contents

5 Causal Inference with Experiments . . . . . . . . . . . . . . . . . . 175

Notation for Causal Inference 177
Randomized Controlled Trials 179
Regression Adjustment 184
Regression Discontinuity Designs 192
Instrumental Variables 199
Design of Experiments 208

6 Causal Inference with Controls. . . . . . . . . . . . . . . . . . . . . . 215

Conditional Ignorability 216
Double Machine Learning 224
Heterogeneous Treatment Effects 231
Using Time Series as Controls 241

7 Trees and Forests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 259

Decision Trees 261
Random Forests 268
Causal Inference with Random Forests 275
Distributed Computing for Random Forests 277

8 Factor Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 282

Clustering 283
Factor Models and PCA 290
Factor Regression 302
Partial Least Squares 308

9 Text as Data. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 316

Tokenization 317
Text Regression 325
Topic Models 329
Word Embedding 338

10 Deep Learning. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 345

The Ingredients of Deep Learning 346
Working with Deep Learning Frameworks 352
Stochastic Gradient Descent 367
The State of the Art 374
Intelligent Automation 380

ISTUDY
 Contents    ix

Appendix: R Primer. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 383

Getting Started with R 384
Working with Data 396
Advanced Topics for Functions 408
Organizing Code, Saving Work, and Creating Reports 414

Bibliography 419
Glossary 424
Acronyms 433
Index 435

ISTUDY
 PREFACE

What Is This Book About?

The practice of data analytics is changing and modernizing. Innovations in computation and
machine learning are creating new opportunities for the data analyst: exposing previously
unexplored data to scientific analysis, scaling tasks through automation, and allowing deeper
and more accurate modeling. Spreadsheet models and pivot tables are being replaced by code
scripts in languages like R and Python. There has been massive growth in digitized information,
accompanied by development of systems for storage and analysis of this data. There has also
been an intellectual convergence across fields—machine learning and computer science, sta-
tistics, and social sciences and economics—that has raised the breadth and quality of applied
analysis everywhere. This is the data science approach to analytics, and it allows leaders to go
deeper than ever to understand their operations, products, and customers.
This book is a primer for those who want to gain the skills to use data science to help make
decisions in business and beyond. The modern business analyst uses tools from machine learn-
ing, economics, and statistics to not only track what has happened but predict the future for
their businesses. Analysts may need to identify the variables important for business policy, run
an experiment to measure these variables, and mine social media for information about public
response to policy changes. A company might seek to connect small changes in a recommen-
dation system to changes in customer experience and use this information to estimate a demand
curve. And any analysis will need to scale to companywide data, be repeatable in the future,
and quantify uncertainty about the model estimates and conclusions.
This book focuses on business and economic applications, and we expect that our core
audience will be looking to apply these tools as data scientists and analysts inside companies.
But we also cover examples from health care and other domains, and the practical material that
you learn in this book applies far beyond any narrow set of business problems.
This is not a book about one of machine learning, economics, or statistics. Rather, this
book pulls from all of these fields to build a toolset for modern business analytics. The material
in this book is designed to be useful for decision making. Detecting patterns in past data can
be useful—we will cover a number of pattern recognition topics—but the necessary analysis
for deeper business problems is about why things happen rather than what has happened. For
this reason, this book will spend the time to move beyond correlation to causal analysis. This
material is closer to economics than to the mainstream of data science, which should help you
have a bigger practical impact through your work.
We can’t cover everything here. This is not an encyclopedia of data analysis. Indeed, for
continuing study, there are a number of excellent books covering different areas of contempo-
rary machine learning and data science. For example, Hastie et al. (2009) is a comprehensive
modern statistics reference and James et al. (2021) is a less advanced text from a similar view-
point. You can view this current text as a stepping stone to a career of continued exploration
and learning in statistics and machine learning. We want you to leave with a set of best prac-
tices that make you confident in what to trust, how to use it, and how to learn more.
x

ISTUDY
 GUIDED TOUR

This book is based on the Business Data Science text by Taddy (2019), which was itself developed
as part of the MBA data science curriculum at the University of Chicago Booth School of Business.
This new adaptation creates a more accessible and course-ready textbook, and includes a major
expansion of the examples and content (plus an appendix tutorial on computing with R). Visit Con-
nect for digital assignments, code, datasets, and additional resources.

Practical Data Science for Decision Making

Our target readership is anyone who wants to get the skills to use modern large-scale data to
make decisions, whether they are working in business, government, science, or anywhere else.
In the past 10 years, we’ve observed the growth of a class of generalists who can understand busi-
ness problems and also dive into the (big) data and run their own analyses. There is a massive demand
for people with these capabilities, and this book is our attempt to help grow more of these sorts of
people. You may be reading this book from a quantitative undergraduate course, as part of your MBA
degree at a business school, or in a data science or other graduate program. Or, you may just be reading
the book on your own accord. As data analysis has become more crucial and exciting, we are seeing
a boom in people switching into data analysis careers from a wide variety of backgrounds. Those
self-learners and career-switchers are as much our audience here as students in a classroom.
All of this said, this is not an easy book. We have tried to avoid explanations that require
calculus or advanced linear algebra, but you will find the book a tough slog if you do not have a
solid foundation in first-year mathematics and probability. Since the book includes a breadth of
material that spans a range of complexity, we begin each chapter with a summary that outlines
each section and indicates their difficulty according to a ski-hill scale:
The easiest material, requiring familiarity with some transformations like logarithms
and exponents, and an understanding of the basics of probability.
Moderately difficult material, involving more advanced ideas from probability and
statistics or ideas that are going to be difficult to intuit without some linear algebra.
The really tough stuff, involving more complex modeling ideas (and notation) and
tools from linear algebra and optimization.
The black diamond material is not necessary for understanding future green or blue sections,
and so instructors may wish to set their courses to cover the easy and moderately difficult sec-
tions while selecting topics from the hardest sections.
The book is designed to be self-contained, such that you can start with little prerequisite
background in data science and learn as you go. However, the pace of content on introductory
probability and statistics and regression is such that you may struggle if this is your first-ever
course on these ideas. If you find this to be the case, we recommend spending some time work-
ing through a dedicated introductory statistics book to build some of this understanding before
diving into the more advanced data science material.

ISTUDY
 xii    Guided Tour

It is also important to recognize that data science can be learned only by doing. This means
writing the code to run analysis routines on really messy data. We will use the R scripting lan-
guage for all of our examples. All example code and data is available online, and one of the
most important skills you will get out of this book will be an advanced education in this pow-
erful and widely used statistical software. For those who are completely new to R, we have also
included an extensive R primer. The skills you learn here will also prepare you well for learning
how to program in other languages, such as Python, which you will likely encounter in your
business analysis career.
This is a book about how to do modern business analytics. We will lay out a set of core
principles and best practices that come from statistics, machine learning, and economics. You
will be working through many real data analysis examples as you learn by doing. It is a book
designed to prepare scientists, engineers, and business professionals to use data science to
improve their decisions.

An Introductory Example
Before diving into the core material, we will work through a simple finance example to illus-
trate the difference between data processing or description and a deeper business analysis.
Consider the graph in Figure 0.1. This shows seven years of monthly returns for stocks in the
S&P 500 index (a return is the difference between the current and previous price divided by
the prior value). Each line ranging from bright yellow to dark red denotes an individual stock’s
return series. Their weighted average—the value of the S&P 500—is marked with a bold line.
Returns on three-month U.S. treasury bills are in gray.
This is a fancy plot. It looks cool, with lots of different lines. It is the sort of plot that you
might see on a computer screen in a TV ad for some online brokerage platform. If only I had
that information, I’d be rich!

S&P500
0.5
Return
0.0
–0.5

2010 2011 2012 2013 2014 2015 2016 2017

FIGURE 0.1 A fancy plot: monthly stock returns for members of the S&P 500 and their average (the bold
line). What can you learn?

ISTUDY
 Guided Tour    xiii

But what can you actually learn from Figure 0.1? You can see that returns do tend to
bounce around near zero (although the long-term average is reliably much greater than zero).
You can also pick out periods of higher volatility (variance) where the S&P 500 changes more
from month to month and the individual stock returns around it are more dispersed. That’s
about it. You don’t learn why these periods are more volatile or when they will occur in the
future. More important, you can’t pull out useful information about any individual stock. There
is a ton of data on the graph but little useful information.
Instead of plotting raw data, let’s consider a simple market model that relates individual
stock returns to the market average. The capital asset pricing model (CAPM) regresses the
returns of an individual asset onto a measure of overall market returns, as shown here:
rjt = αj + βjmt + εjt (0.1)
The output rjt is equity j return at time t. The input mt is a measure of the average return—the
“market”—at time t. We take mt as the return on the S&P 500 index that weights 500 large
companies according to their market capitalization (the total value of their stock). Finally, εjt is
an error that has mean zero and is uncorrelated with the market.
Equation (0.1) is the first regression model in this book. You’ll see many more. This is a
simple linear regression that should be familiar to most readers. The Greek letters define a line
relating each individual equity return to the market, as shown in Figure 0.2. A small βj, near zero,
indicates an asset with low market sensitivity. In the extreme, fixed-income assets like treasury
bills have βj = 0. On the other hand, a βj > 1 indicates a stock that is more volatile than the mar-
ket, typically meaning growth and higher-risk stocks. The αj is free money: assets with αj > 0 are
adding value regardless of wider market movements, and those with αj < 0 destroy value.
Figure 0.3 represents each stock “ticker” in the two-dimensional space implied by the mar-
ket model’s fit on the seven years of data in Figure 0.1. The tickers are sized proportional to
each firm’s market capitalization. The two CAPM parameters—[α, β]—tell you a huge amount
about the behavior and performance of individual assets. This picture immediately allows you
to assess market sensitivity and arbitrage opportunities. For example, the big tech stocks of
Facebook (FB), Amazon (AMZN), Apple (AAPL), Microsoft (MSFT), and Google (GOOGL)
all have market sensitivity β values close to one. However, Facebook, Amazon, and Apple
generated more money independent of the market over this time period compared to Micro-
soft and Google (which have nearly identical α values and are overlapped on the plot). Note
0.3
Equity return
0.1
−0.1
−0.3

−0.2 −0.1 0.0 0.1 0.2

Market return

FIGURE 0.2 A scatterplot of a single stock’s returns against market returns, with the fitted regression
line for the model of Equation (0.1) shown in red.

ISTUDY
 xiv    Guided Tour

FB

0.015
AMZN

AAPL
0.010

V
Alpha

BA
AMGN
DIS
0.005

T
PEP JNJ PFE GOOG
MSFT
WFC GE
PG KO
WMT JPM
0.000

CVX
IBM XOM
ORCL BAC
CSCO

0.4 0.6 0.8 1.0 1.2 1.4 1.6

Beta

FIGURE 0.3 Stocks positioned according to their fitted market model, where α is money you make
regardless of what the market does and β summarizes sensitivity to market movements. The tickers are sized
proportional to market capitalization. Production change alpha to α and beta to β in the plot axis labels.

that Facebook’s CAPM parameters are estimated from a shorter time period, since it did not
have its IPO until May of 2012. Some of the older technology firms, such as Oracle (ORCL),
Cisco (CSCO), and IBM, appear to have destroyed value over this period (negative alpha).
Such information can be used to build portfolios that maximize mean returns and minimize
variance in the face of uncertain future market conditions. It can also be used in strategies
like pairs-trading where you find two stocks with similar betas and buy the higher alpha while
“shorting” the other.
CAPM is an old tool in financial analysis, but it serves as a great illustration of what to strive
toward in practical data science. An interpretable model translates raw data into information that
is directly relevant to decision making. The challenge in data science is that the data you’ll be
working with will be larger and less structured (e.g., it will include text and image data). Moreover,
CAPM is derived from assumptions of efficient market theory, and in many applications you won’t
have such a convenient simplifying framework on hand. But the basic principles remain the same:
you want to turn raw data into useful information that has direct relevance to business policy.

Machine Learning
Machine learning (ML) is the field of using algorithms to automatically detect and predict pat-
terns in complex data. The rise of machine learning is a major driver behind data science and a
big part of what differentiates today’s analyses from those of the past. ML is closely related to
modern statistics, and indeed many of the best ideas in ML have come from statisticians. But
whereas statisticians have often focused on model inference—on understanding the parameters
of their models (e.g., testing on individual coefficients in a regression)—the ML community
has historically been more focused on the single goal of maximizing predictive performance
(i.e., predicting future values of some response of interest, like sales or prices).

ISTUDY
Other documents randomly have
different content
 I went to the hillside, I went to pray, I
know the angels done changed my name, Done
changed my name for the coming day, Thank
God the angels done changed my name.

You May Bury Me in the East.

You may bury me in the East, You may

bury me in the West, But I'll hear the trumpet sound in that
morning. In that morning, my Lord
How I long to go for, For to hear the trumpet sound, In that
morning.

It is often stated that the chief rhythmic characteristic of the negro

music is the so-called "Scotch jerk," the jump away from the
normally accented note to another, thrice exemplified in the third line
of "The Angels Done Changed My Name," and imitated in ragtime. A
more typical instance of it is "Didn't My Lord Deliver Daniel" (Figure
XXXV), which also further illustrates major-minor idiom in its
constant see-saw between G minor and B-flat major. It is pointed
out that the slaves had a strong sense of time, that the
overwhelming majority of their songs are in duple or march time,
with very few in the more graceful but less vehement triple measure,
and that in their "shouts" or religious dances they rocked themselves
into paroxysms of rhythmic excitement, one group clapping the
meter while the others sang and scuffled with a "jerking, twitching
motion which agitated the entire shouter and soon brought out
streams of perspiration."[66] No doubt the jerk evidences their love
of strong accentuation; but it must be noted that accentuation is a
purely local thing, affects the meter rather than the rhythm, and
may be assumed and put off by a tune (as in the "ragging" of a
standard melody) without changing its essential curve.

Figure XXXV.
Didn't My Lord Deliver Daniel?

Didn't my Lord deliver Daniel D'liver

Daniel, d'liver Daniel, Didn't my Lord deliver
 Daniel, And why not a ev'ry man?

Going Up.

Oh, yes, I'm going up, going up,

going all the way, Lord, going up,
going up, to see the heavenly land.

Far more significant, therefore, than their half-barbaric fondness for

the jerk is the grasp shown by negroes over the larger and nobler
reaches of rhythm, their feeling for the phrase as a whole and ability
to impress upon it a firm and yet varied profile. The second half of
"You May Bury Me in the East," with its bold festooning of outline,
even more strikingly the tune "Going Up," with its piquant silences
and its even-paced insistence on "going all the way, Lord," show a
unity in their variety, a certain "all-of-a-piece-ness," compared with
which even "Didn't My Lord Deliver Daniel" seems scrappy, and the
ordinary ragtime effusion pitifully poverty-stricken. There is plenty of
internal evidence, too, that these happy results are attributable to
genuine musical imagination, and not to luck in the servile following
of felicitous word-patterns. Indeed, the frequency with which
unimportant words are accented and important ones slurred over
shows that, as is so often the case with great melodists like
Schubert, the words were regarded more or less as convenient pegs
to hang the melodies on, and the specifically musical faculty did not
easily brook interference. "The negroes keep exquisite time," writes
one of the editors of "Slave Songs in the United States," the best of
the collections, "and do not suffer themselves to be daunted by any
obstacle in the words. The most obstinate hymns they will force to
do duty with any tune they please, and will dash heroically through a
trochaic tune at the head of a column of iambs with wonderful skill."
The sense of independent tone-pattern, which when possessed by
individual geniuses in supreme degree gives us the immortal
melodies of a Beethoven or a Brahms, waxes and wanes in these
childlike tunes, sometimes falling back into platitude, but sometimes
advancing to a real distinction and beauty.
Whether this beauty is of the kind we have desiderated as the
highest quality folk-song can have, rendering it "suitable to enter
into music that may bear comparison with the best music of the
world," is a further question, and one which brings us at length to
the highly controversial matter of the kind of treatment that the
composer should give folk-material in incorporating it into his more
finished art. The variations of taste concerned here are so subtle
that probably unanimity of judgement, even if it be desirable, will
never be attained. Yet it is certain that treatment of some sort there
must be. The mere collecting, collating, and setting forth of folk-
songs, attractively arranged for instruments or even orchestrated,
such as we have seen much of from all countries in recent years, is
no more musical art than a pile of bricks is a building, or a series of
anecdotes literature. So far as it tends to content the public with
such potpourris, the fad for folk-song is positively injurious to taste,
in something the same way that our modern floods of petty
journalism are injurious to the capacity for sustained reading.
Moreover, even on their own level such medleys are apt to be
unsatisfactory; for the tunes themselves are so definite, brief, and
complete, and the transitional passages between them are therefore
so obtrusively transitional, that the net effect is that of the ill-baked
bread pudding from which we eat nothing but the raisins. Mr.
Coleridge-Taylor's "Twenty-Four Negro Melodies," despite incidental
attractions, are on the whole an example of this bad model.
Far worse, however, are those "improvements" of folk-song which
consist in a general prettifying of its homely simplicity with all the
refinements and luxuries of sophisticated musical technique—as if a
country maiden should conceal her healthy color under layers of
rouge. Strange that composers skilful enough to use them should
not recognize the inappropriateness of Wagnerian chromatics and
Debussyan whole-tone scale harmonies, to say nothing of all sorts of
rich dissonantal trappings, to tunes as diatonic as "God Save the
King" and as square cut as the "Hymn of Joy." One would think that
the sense of humor, which revels in incongruity in music as in other
things, would keep them from doing it and us from taking it so
seriously. It would be invidious to name examples, but they can be
discovered by the discerning; for not even the negro complexion is
proof against this brand of talcum powder.
The kind of change that is both legitimate and necessary may
perhaps be best suggested by another example, "Deep River." Here
we have, in the first phrase, that free and firm molding of rhythmic
pattern which is often so surprising in these songs, so that we might
look far in the best composers without finding its peer in deliberate,
calm beauty. But just as our hearts are responding to the wave of
emotion thus generated it strikes, so to speak, a dead wall, falls
shattered, and has to begin over again, without being able to
recover the lost momentum. The imagination is vital as far as it
goes, but its span is short, it lacks sustained power and cumulative
force. What is needed in the composer who would deal with such
material, then, in addition to a tact that enters into its spirit, is a
synthetic imagination capable of rounding out its incompleteness, of
tracing the whole of the curve it suggests, of developing into full life
what it presents only as a germ.
 Figure XXXVI.
Deep River.

Deep river, my home is over Jordan,

Deep river, Lord I want to cross over in to camp ground.

How difficult such a truly creative treatment is, only those fully know
who have tried it; how rare, musical literature testifies. To add a
measure to a folk-song is almost like adding a cubit to one's stature,
and for the same reason—that addition is not what is requisite, but
organic growth. That it is possible we see in Brahms's masterly
treatment of German student songs in his "Academic Festival
Overture"; that it can be applied to negro melodies we have been
shown especially by Dvořák. In his "New World" symphony and his
"American" Quartet and Quintet he assimilated a peculiar idiom so
perfectly that there is not a note, even in the highly complex
harmonies toward the end of the symphony, that does not take its
place in the scheme unobtrusively. While the harmonic idiom
preponderantly of simple triads dictated by the material is
maintained with an unerring sense of style, these commonest of all
chords are so deftly managed that they never become
commonplace. The twin pitfalls of platitude and sophistication are
avoided with equal success. The same felicity is attained in the
construction. However brief the themes, they do not sound trivial or
unconvincing, because we feel they have reached their natural
growth. Above all, the same sympathy and power that are shown in
these technical matters so control the conception as a whole that
these works form a true idealization of negro feeling, in its moods
both of half-barbaric dance and of naïvely pathetic sentiment.
Dvořák's example suffices by itself, then, to show that the negro
music, in the hands of a master, is capable of two of the three
qualities we demanded of any folk-song—idiomatic distinctiveness
and capacity for organized beauty. Does he also demonstrate in it
the third—adequacy to interpret the American temper? Something
closely kindred to that temper and easily endeared to it there
certainly is in the restless rhythmic energy, the unceasing motion
and quick changes of these scherzos, the vigor and dispatch of these
allegro movements. Like similar syncopations and other rhythmic
peculiarities that we find in those of our composers who have more
than their share of our national nervous energy, such as Chadwick
and Whiting, the negro rhythms have a crispness and buoyancy that
is somehow appropriate to our clear skies and self-helpful society.
They give at least a far fairer portrait of us than the caricature of
ragtime. In its more sentimental moods, too, negro music has an
unsophistication, an unreserved naïveté, that reminds us of similar
traits in the traditional conception of our fellow countrymen. It thus
seems to express more of our national temperament, and to leave
less of it unexpressed than would on the whole any other body of
folk-song.
Yet the very attempt to formulate these considerations forces us to
realize how hopelessly inadequate they are as an account of the
possibilities of America in music. The picture they give of the
national type may do something like justice to it as it existed in
earlier times and simpler surroundings, as it appears, for instance, in
the pages of Mark Twain or Bret Harte, and as it is symbolised in the
person of Uncle Sam; but the modern American is a being quite
other, far more complex, far more cosmopolitan, the American not of
nineteenth century New England but of the twentieth century
"melting pot." He is wholly incommensurate not only with negro
music or any folk music, but with even individual composers like
Dvořák in whom emotion far outruns intellectual subtlety. No folk
music, let us repeat, no individual composer, no school of
composers, can "express" America. The age of such simplicities is
past, if it ever existed. Whether we like it or not, we have to take
our age and our country as they are; they are an age of rapidly
accelerating intercommunication of all peoples and a country in
which the internationalism that thus slowly results is being hastened
by actual admixture on a heretofore unprecedented scale. Such a
condition doubtless has its bad as well as its good aspects; but if
those who bemoan our "featureless cosmopolitanism" and advocate
an impossible parochialism as the only remedy would try rather to
see how a wider outlook and a larger sympathy may deepen our art
and make it more truly human by laying less stress on local,
national, or even racial types, and more on the untrammeled
expression of the greatest possible variety of individuals, music
would fare better. "National literature:" wrote Goethe to Eckerman in
1827, "the term has no longer much meaning to-day; the time for
universal literature is come, and each ought to work to hasten its
advent." Signs are not wanting that the condition thus discerned by
the wisest men a century ago is now gradually getting itself
acknowledged in general practice.

V
If we accept, in the light of the foregoing considerations, the ideal of
enlightened eclecticism, not only for our own music here in America
but measurably for all modern music, since it is all subject to the
internationalization so characteristic of our time, the chief
undertaking that remains to us will be an attempt to define the
position of the American composer in relation to such eclecticism,
the advantages and disadvantages of his situation, the pitfalls he
must avoid, and the opportunities he should embrace. From this
point of view it will be seen that the enthusiasts of nationalism, in
advising our composer to confine himself to Indian, Negro, or
ragtime material, in adjuring him not to listen to the siren voice of
Europe, are not merely misleading but cheating him. They are asking
him to throw away his birthright of wide cosmopolitan influence for a
mess of purely parochial pottage. They are bewailing the lack in
America of just those geographical and racial boundary lines that
split up Europe into a series of more or less petty and hostile camps.
They are inviting us to descend from the point of vantage good
fortune has given us, a little removed both in space and in time from
the thick of the battle.
For it is indeed the peculiar good fortune of the young American
composer that he finds spread out before him, as the models
through the study of which he is to acquire an important part of his
technical equipment and of his general attitude towards art, the
masterpieces of the various European countries, among which he
may pick and choose as his individual taste directs, and without
being hampered by those annoying racial and national jealousies
from which the most intelligent European cannot quite free himself.
What he may acquire of the special virtue of each school—the
delicacy and distinction of the French, the solid structural power of
the German, the suave and rich coloring of the Russian, the austere
dignity of the English—is limited, not by the accident of birth, but
only by his own assimilative power. No element in his complex
nature need be starved for want of its proper food. He is placed in
the midst of the stream of world influences to make of himself what
he will and can.
Is it not inconceivable that one thus privileged to speak, within the
measure of his ability, a world language should ever content himself
with a Negro or Indian dialect? It would be so perhaps did we not
consider that, in order to speak the world language of cosmopolitan
music as it exists to-day, one must spend years in laborious
discipline and in obscurity, while any tyro can make a certain effect
and gain a certain prominence by stammering in an idiom strongly
enough tinctured with local color. Vanity is the immemorial enemy of
art; if the itch to be conspicuous once infect him, the artist forgets
all those subtle adaptations, those difficult reconcilements, which
were formerly his passion, and makes a crude effect that appeals
much more to the primitive minds of the masses. And this he may
do quite unconsciously and in the sincere belief that he is pursuing
the highest ideals. In the presence of the immediate good, of
recognition and acclaim, it is pitifully easy to forget the remote
better, the broader, finer, subtler beauty that is not yet understood.
But if the picturesque, the quaint, the piquant, is by nature more
quick to appeal than the beautiful, it is also more short-lived. For this
reason those writers in all ages and countries who depend largely on
local color are promptly acclaimed and soon forgotten, while those
who aim at the more universal human qualities win gradually a place
that proves permanent. Bret Harte was doubtless considered more
"American" by his own generation than Emerson. Shakespeare is far
less English than Defoe, Dante is not so notably Italian, or Goethe so
notably German, as are many lesser men. Or, to come back to
music, where are now the Russian "nationalists" who excluded
Tschaikowsky the "cosmopolite" from their magic circle? For a while
we listened to their melancholy Russian cadences with fascinated
interest, in spite of their crude harmonization, their incoherent form,
their lack of instinct for style, because we were pleased with the
novelty. Now the novelty has worn off, and for human nature's daily
food we find Tschaikowsky, who made the most of his opportunities,
rose above a narrow exclusiveness, and assimilated power wherever
he found it, far preferable.
The true difficulty of the American composer's position, then, is to
be found, not in the poverty of the native folk-song, but in the
confusing variety of the foreign influences in which he is so rich. He
has suffered and is still suffering from an embarrassment of riches,
from a mental indigestion. His cosmopolitanism is indeed too often
"featureless," and his readiness to be influenced an evidence of
weakness rather than strength, a flat rather than a broad
eclecticism. His technique is miscellaneous, his style without
distinction, his art as a whole lacks individuality. This featurelessness
is the typical defect of American compositions of the present
generation, perhaps—typical in spite of some notable exceptions.
The technical deficiencies of our pioneer forefathers are more and
more becoming things of the past; free intercourse with Europe and
the wholesale importation of skilled European musicians have refined
away the crudities with surprising rapidity; there are among us to-
day musical workmen whose skill in symphony, chamber music, and
opera will compare favorably with that of Europeans. Where we still
fail is in that subtle, indefinable, and indispensable touch of personal
distinction which may be recognized in artists so diverse, both
individually and racially, as Strauss, d'Indy, Debussy, Rachmaninoff,
Paderewski, Sibelius, Elgar. What is the secret of this distinction?
We may get a clue to the right answer by considering a peculiar
case, an exceptional case, among ourselves—the exception that
proves the rule—the case of MacDowell. The supreme place he
undoubtedly holds among our composers is due precisely to this
quality of personal distinction, of individuality, to the fact that his
music, in spite of the pronounced Grieg and Raff elements in it, does
not sound quite like that of any one else. Technically MacDowell has
grave deficiencies; his harmonic system is singularly limited,
mannered, and monotonous; his polyphony is weak; his "drawing,"
as a painter would say, is often halting and awkward. His range of
expression, moreover, is not wide, and within it he frequently cloys
by an over-sweet sentimentalism. But MacDowell is sincere, and he
is always himself. There are no unfused elements in his style, no
outstanding features, that we feel to have been borrowed and not
assimilated. His style is very narrow, but it is his own. And the result
is that, although we shall soon forget some of our composers who
are far cleverer than he, we shall not forget MacDowell.
The enemies of eclecticism have thus expressed a half-truth, we
begin to see, when they call it flabby. Only too easily does it become
so. As dangerous as it is desirable, it will contribute to the formation
of an artist only when it is controlled by an instinctive sense of how
much one can assimilate, and the courage to reject the rest. And
here we come to one of those peculiar difficulties of the position in
which the American composer finds himself. It is hard for him,
recognizing, as his natural alertness of perception and his detached
point of view enable him to do, the merits of many different
European aims and methods, and, mainly sensitive as he must be to
his own shortcomings in respect to any of them—it is hard for him to
distinguish between those that he can possibly assimilate to his own
uses and those that must remain alien to him; and it is doubly hard
to let the latter alone, voluntarily restricting his field in order that he
may be the master of it. Yet these selections, these sacrifices, are at
the very foundation of artistic personality. It is no more possible for
a human being to be, let us say, at once as subtle as Debussy and
as gorgeous as Strauss than it is to be in two places at once. Which
will you do without? But the young American composer is at once
too timid and too ambitious to do without anything; in the attempt
to be everywhere at the same time he cuts himself up into little
pieces that end by being nowhere.
The frank and courageous acceptance of limitations is, in truth, the
first step toward artistic individuality; a man can never be an
individual, as the very derivation of the word may remind him, so
long as he remains divided, spread out very wide and very thin,
unwilling to take sides, but only when he concentrates himself, is
loyal to one cause, grows out from one nucleus. What this nucleus
shall be, indeed, differs according to circumstances. For the
European musician it is to some extent decided beforehand, by the
conditions of birth, of national and racial allegiance. The American,
as we said, to begin with is freer in this respect; but we may now
add that he is no less bound to find a cause, a unifying center, if he
would get beyond mere clever imitation and become a genuine
person. He must love his cause so singly that he will cleave to it, and
forsake all else. Now what is this cause for the American composer
but the utmost musical beauty that he, as an individual man with his
own qualities and defects, is capable of understanding and striving
towards? And what is the "all else" that he must forsake, save those
types of musical beauty which, whatever may be their intrinsic
worth, do not come home to him, do not arouse a sympathetic
vibration in him, leave him cold? He must take sides. He must be,
not a philosopher, but a partisan. He must have good hearty
enthusiasms, and good hearty prejudices. Only so can he be an
individual.
It matters not one jot, provided this course of personal loyalty to a
cause be steadfastly pursued, what the special characteristics of
style of the music may be to which one gives one's devotion. Let A
give his life to studying the delicate color scheme of the French
"ultra-moderns"; let B find his joy in a polyphony based on Bach's;
let C develop lovingly the cadences or rhythms of Negro and Indian
tunes; all three will be good musicians, all three good Americans—
for, after all, American music is only music. The man who is neither
good musician nor good American is the botcher, the dilettante, the
clever amateur—he who is too lazy to learn his business, too
pretentious to limit his claims, too busy talking about art to study it.
Such babblers have always been, and always will be, naturally, far
more numerous than the efficient workers; and they will doubtless
continue to fill the newspapers and magazines with their silly
superficialities, and do their utmost to confuse the public into
forgetting that sincerity and skill are the only things that can ever be
justly demanded of an artist.

VI
In demanding skill and sincerity of our composers, however, we are
requiring of them, as a little analysis will suffice to show, labors and
sacrifices of which only the rarest natures are capable; and it may
well be that the unsatisfactory character of composition in America is
due far more to the rarity of men able or willing to undertake such
labors and endure such sacrifices than to the difficulties of the
æsthetic problems we have so far been considering. Let us, then, in
closing, try to suggest answers to the purely practical questions: Is
there anything about our social and economic system that lays
especial burdens on creative artists? If there is, is there any hope of
correcting it? Whether it may be corrected or not, may our
composers, through candid recognition of it, be saved from
dissipation of energy and helped to concentrate their efforts on
objects most likely to be achieved, and most worth achieving? We
shall answer all these questions affirmatively.
First we must note that the amount and intensity of mental
application involved in composition is something of which the layman
has little idea. The technique to be mastered by the composer is
singularly difficult; the tonal material he works in is subtle and
intangible; its relationships, harmonic, rhythmic, melodic,
polyphonic, which he must learn not only to understand but to
manipulate, are of an indescribable complexity; and he achieves
command of all these fundamental or grammatical means of his art
only to face the far subtler distinctions of structure and style on a
wise apprehension of which depends his artistic individuality.
Moreover, if he would take advantage of the wide and unbiased view
of European music which we have seen to be a special privilege of
the intelligent American, he must do far more than hear or read the
chief works of many masters; he must know them in and out, must
learn to breathe the peculiar atmosphere of each,—must, in short,
live with them. And more than that, for after analysis comes
synthesis, after assimilation, creation; and as the one requires
laborious, minute, detailed study, the other requires a wide margin
of leisure in which the mind can forget all these details, empty itself
of all irrelevance, and prepare to receive whatever thoughts may
visit it. Here is more time needed, in great spaces. This is a full and
varied way of living, indeed, that we are sketching; and we have not
yet made out how the artist is to live at all. How is he to get money
to support himself?
Not, certainly, from his compositions. They will do well if they bring
him enough to pay for ink and paper; they will surely not pay for
their own copying. "A man," says Mr. Graham Wallas,[67] "who gives
the best strength of each day to dreaming about the nature of God
or the State, or the shape of the earth, or the relation of the sides of
a triangle to its hypotenuse, produces nothing which at the end of
the day he can easily sell. Since the actual process of inference is
unconscious, and his voluntary control over it indirect and uncertain,
he is not even sure that he will produce any result at all, whether
salable or unsalable, by months of effort. How then shall he live?" If
this is the situation of the creative thinker in science, what shall we
say of that of the creative thinker in art? As we have seen in
discussing the relations of democracy and music, the class which in
the eighteenth century bought the wares of the composer finds its
analogue, under our capitalistic industrial system, in the frivolous
plutocracy, who demand of music curiosities, novelties, and
entertainment. The vast mass of listeners emerging from below, on
the other hand, of crude and childlike taste, prefer stories (program
music), day-dreaming, and sensationalism to beauty. Confronted by
these two classes, the composer will find his sincerity likely to cost
him dear. If he is really sincere, if he is trying to write music that
presents the kind of beauty that he hears, and that no one else has
heard in just that way before, he will find himself enjoying it in a
minority of one. Yet the alternative, to prostitute himself and "give
the public what it wants," is even worse; and when the public says
to him, in the words of Mozart's publisher, "Write in a more easy,
popular style, or I will not print a note or give you a kreuzer," his
answer can be no other than Mozart's: "Then, my good sir, I have
only to resign myself and die of hunger."
Or rather, and here is the special irony of the situation, his
alternative is not a literal physical hunger, but that subtler hunger
that follows the denial of the imperious instinct to create beauty; he
has not to starve his body of bread, but his soul of music. For while
society withholds with one hand, so to speak, any payment for the
best work he can do, because it is too good, because it requires too
long to be understood, it freely offers him with the other a bare
livelihood, if not more, for work of secondary value—teaching,
performance, exposition, anything but creation. It constantly pulls,
pushes, cajoles, persuades, coaxes, browbeats him from the
superior to the inferior activity. It so fills his days with the one that
even if at long intervals an opportunity for the other presents itself
he has hardly the spirit to seize it. It deadens him with detail, drugs
him with drudgery, cages him until he forgets how to sing. Where,
as in America, there exists a very "high standard of living," as it is
quaintly called, meaning that many and costly material wants have
to be met before spiritual needs can be considered, the labor
imposed by such a struggle may be overwhelming. And it is
superimposed, we must remember, on the other labor, the creative
one, described above. The same nerves, body, and brain, in the
same twenty-four hours each day, must sustain the two labors, one
to earn a livelihood, the other to make use of it. No wonder few can
endure it; no wonder most give it up in despair or dull indifference,
and content themselves with the livelihood without taking the
trouble to live.
Not only, moreover, are the broad facts of economics, under a
capitalistic-industrial system, thus flatly inimical to creative work, but
in a plutocratic civilization like ours the more subtle forces of public
opinion are perhaps even more fatal to it, because more pervasive
and intangible. In Europe the impecunious artist is accepted with
tolerance, even with a touch of respect, and suffered to live
undisturbed in his Bohemia and to pursue his dreams. To us, who
still as a people recognize no measure of achievement but income,
and who accept without a murmur the domination of mass-
convention in most matters of opinion, he is something worse than
an interesting eccentric or even a harmless crank; he is something of
a sybarite and a skulker; he is one who "doesn't play the game."
Therefore he need look to us for understanding or sympathy no
more than for more material rewards. If he wishes to be approved
of, let him do something useful—that is, something that pays.
When we realize the penalties that are thus piled upon the head of
the artist whose only offense is that he wishes to give something to
society of which it does not yet recognize the value, our wonder that
there is so little American composition of the first quality changes to
surprise that there is any. We begin to suspect—as Ruskin did at
forty, and devoted the rest of his life to demonstrating the truth of
his suspicion—that the decadence of art we witness all around us is
only a symptom of a deeper disease, and that, as William Morris
expressed it, "Slavery lies between us and art."[68] Capitalistic
industrialism, as Matthew Arnold saw, "materializes our upper class,
vulgarizes our middle class, brutalizes our lower class"; and under
such conditions vital art can have no secure or assured life. It may
well be, therefore, that art can only in the long run be saved, like
society itself, by the fairer, freer, humaner system that socialism
promises. It cannot but thrill all true lovers of art to find its claims,
with those of a liberalized society, being championed to-day, no
longer merely by individual thinkers like Ruskin and Morris, but by
great representative bodies like the British Labor Party. "Society, like
the individual," says a draft report of this party[69] "does not live by
bread alone—does not exist only for perpetual wealth production.
The Labour Party will insist on greatly increased public provision
being made for scientific investigation and original research, in every
branch of knowledge, not to say also for the promotion of music,
literature, and fine art, which have been under capitalism so greatly
neglected, and upon which, so the Labour Party holds, any real
development of civilization fundamentally depends."
Finally, however, inspiring as are the hopes these words suggest, the
American composer need not await their realization before putting
forth those individual efforts without the aid of which, after all, they
can never attain it. Music, like society, has reached its present state
only through the struggles, against immense odds, of its martyrs and
its heroes: not only of Bach, of Mozart, of Beethoven, of Schubert, of
Wagner, of Brahms, but of countless others who have wrought and
suffered in obscurity and with a consecration of their more limited
powers to the great cause of beauty. And if American life lays almost
crushing burdens on artistic initiative, there is also in the best
American tradition a courage, an independence, a certain nonchalant
and plucky self-reliance that ought to carry an artist far on the
solitary path he has to travel. It ought to keep him from turning
back, though it could not guard him against wandering and getting
lost. All that can help him there is clearsightedness, a realistic and
unsentimental view of the society in which he lives and the terms on
which he lives in it. He must discharge the work he does for a
livelihood as conscientiously as he can, but meanwhile not forget to
live also. He must not make the tragic mistake, the unpardonable sin
of the artist, described by Thoreau: "To please our friends and
relatives we turn out our silver ore in cartloads, while we neglect to
work our mines of gold known only to ourselves, far up in the
Sierras, where we pulled up a bush in our mountain walk, and saw
the glittering treasure. Let us return thither. Let it be the price of our
freedom to make that known." He must cut down his material
requirements to the minimum and honor his own poverty. He must
learn to find his satisfaction in the work itself, and not expect
recognition, which is bound to be late (even later in America than
elsewhere), and likely to be mistaken. Above all, he must not pity
himself or grow embittered, for in the possession of a lifelong
enthusiasm, an ideal that he can always work towards and will never
reach, he has the best gift that life has to offer.
 FOOTNOTES:

[59] The New Republic, October 16, 1915.

[60] "Two Views of Ragtime." The Seven Arts, July, 1917.
[61] The time is really 4-8, though marked 2-4.
[62] The Times, London, February 8, 1913, quoted in Boston
Symphony Orchestra Program Books, vol. 32, p. 1186.
[63] See, for instance, Mr. Carl van Vechten's "Interpreters and
Interpretations."
[64] Quoted by Mr. Charles L. Buchanan in an admirably sane
article on "Rag Time and American Music" in The Opera
Magazine, February, 1916.
[65] For example:

"They got a fiddler there

That always slickens his hair,
An' folks he sure do pull some bow,"

from "The Memphis Blues," in which Mr. H. K. Moderwell assures

us we shall find "characteristic verse of a high order."
[66] The Nation, May 30, 1867.
[67] "The Great Society," by Graham Wallas.
[68] Quoted in "The Socialist Movement," by J. Ramsay
MacDonald, p. 86.
[69] The Labor Party's Draft Report on Reconstruction: "The Aims
of Labour," by Arthur Henderson, Appendix, page 106.
Printed in the United States of America.