Download the full version of the textbook now at textbookfull.

com

Data Science and Machine Learning

Mathematical and Statistical Methods Chapman
Hall Crc Machine Learning Pattern Recognition
1st Edition Dirk P. Kroese
https://textbookfull.com/product/data-science-and-
machine-learning-mathematical-and-statistical-
methods-chapman-hall-crc-machine-learning-pattern-
recognition-1st-edition-dirk-p-kroese/

Explore and download more textbook at https://textbookfull.com

 Recommended digital products (PDF, EPUB, MOBI) that
you can download immediately if you are interested.

Machine Learning for Factor Investing R Version Chapman

and Hall CRC Financial Mathematics Series 1st Edition
Guillaume Coqueret
https://textbookfull.com/product/machine-learning-for-factor-
investing-r-version-chapman-and-hall-crc-financial-mathematics-
series-1st-edition-guillaume-coqueret/
textbookfull.com

Practical Machine Learning and Image Processing: For

Facial Recognition, Object Detection, and Pattern
Recognition Using Python Himanshu Singh
https://textbookfull.com/product/practical-machine-learning-and-image-
processing-for-facial-recognition-object-detection-and-pattern-
recognition-using-python-himanshu-singh/
textbookfull.com

Statistical Methods for Machine Learning: Discover How to

Transform Data into Knowledge with Python Jason Brownlee

https://textbookfull.com/product/statistical-methods-for-machine-
learning-discover-how-to-transform-data-into-knowledge-with-python-
jason-brownlee/
textbookfull.com

The Holocaust and the Nakba A New Grammar of Trauma and

History Bashair Bashair

https://textbookfull.com/product/the-holocaust-and-the-nakba-a-new-
grammar-of-trauma-and-history-bashair-bashair/

textbookfull.com
Cool Knitting for Kids A Fun and Creative Introduction to
Fiber Art Cool Fiber Art Kuskowski

https://textbookfull.com/product/cool-knitting-for-kids-a-fun-and-
creative-introduction-to-fiber-art-cool-fiber-art-kuskowski/

textbookfull.com

Flourish by Design 1st Edition Nick Dunn

https://textbookfull.com/product/flourish-by-design-1st-edition-nick-
dunn/

textbookfull.com

Solar Chimney Power Plant Generating Technology 1st

Edition Ming

https://textbookfull.com/product/solar-chimney-power-plant-generating-
technology-1st-edition-ming/

textbookfull.com

Automation in the virtual testing of mechanical systems:

theories and implementation techniques First Edition
Haugen
https://textbookfull.com/product/automation-in-the-virtual-testing-of-
mechanical-systems-theories-and-implementation-techniques-first-
edition-haugen/
textbookfull.com

Investment Banking : Concepts Analysis And Cases 3Rd

Edition Pratap Giri

https://textbookfull.com/product/investment-banking-concepts-analysis-
and-cases-3rd-edition-pratap-giri/

textbookfull.com
Indigenous Ways of Knowing in Counseling Theory Research
and Practice Lisa Grayshield

https://textbookfull.com/product/indigenous-ways-of-knowing-in-
counseling-theory-research-and-practice-lisa-grayshield/

textbookfull.com
Data Science and Machine Learning
Mathematical and Statistical Methods
Chapman & Hall/CRC Machine Learning & Pattern
Recognition

Introduction to Machine Learning with Applications in Information

Security
Mark Stamp

A First Course in Machine Learning

Simon Rogers, Mark Girolami

Statistical Reinforcement Learning: Modern Machine Learning

Approaches
Masashi Sugiyama

Sparse Modeling: Theory, Algorithms, and Applications

Irina Rish, Genady Grabarnik

Computational Trust Models and Machine Learning

Xin Liu, Anwitaman Datta, Ee-Peng Lim

Regularization, Optimization, Kernels, and Support Vector Machines

Johan A.K. Suykens, Marco Signoretto, Andreas Argyriou

Machine Learning: An Algorithmic Perspective, Second Edition

Stephen Marsland

Bayesian Programming
Pierre Bessiere, Emmanuel Mazer, Juan Manuel Ahuactzin, Kamel
Mekhnacha

Multilinear Subspace Learning: Dimensionality Reduction of

Multidimensional Data
Haiping Lu, Konstantinos N. Plataniotis, Anastasios Venetsanopoulos
Data Science and Machine Learning: Mathematical and Statistical
Methods
Dirk P. Kroese, Zdravko I. Botev, Thomas Taimre, Radislav Vaisman

For more information on this series please visit: https://www.crcpress.com/Chapman--

HallCRC-Machine-Learning--Pattern-Recognition/book-series/erie
Data Science and Machine Learning
Mathematical and Statistical Methods

Dirk P. Kroese
Zdravko I. Botev
Thomas Taimre
Radislav Vaisman
Front cover image reproduced with permission from J. A. Kroese.

CRC Press
Taylor & Francis Group
6000 Broken Sound Parkway NW, Suite 300
Boca Raton, FL 33487-2742

© 2020 by Taylor & Francis Group, LLC

CRC Press is an imprint of Taylor & Francis Group, an Informa business

No claim to original U.S. Government works

Printed on acid-free paper

International Standard Book Number-13: 978-1-138-49253-0 (Hardback)

This book contains information obtained from authentic and highly regarded sources.
Reasonable efforts have been made to publish reliable data and information, but the author
and publisher cannot assume responsibility for the validity of all materials or the
consequences of their use. The authors and publishers have attempted to trace the copyright
holders of all material reproduced in this publication and apologize to copyright holders if
permission to publish in this form has not been obtained. If any copyright material has not
been acknowledged please write and let us know so we may rectify in any future reprint.

Except as permitted under U.S. Copyright Law, no part of this book may be reprinted,
reproduced, transmitted, or utilized in any form by any electronic, mechanical, or other
means, now known or hereafter invented, including photocopying, microfilming, and
recording, or in any information storage or retrieval system, without written permission
from the publishers.

For permission to photocopy or use material electronically from this work, please access
www.copyright.com (http://www.copyright.com/) or contact the Copyright Clearance
Center, Inc. (CCC), 222 Rosewood Drive, Danvers, MA 01923, 978-750-8400. CCC is a
not-for-profit organization that provides licenses and registration for a variety of users. For
organizations that have been granted a photocopy license by the CCC, a separate system of
payment has been arranged.

Trademark Notice: Product or corporate names may be trademarks or registered

trademarks, and are used only for identification and explanation without intent to infringe.
Visit the Taylor & Francis Web site at
http://www.taylorandfrancis.com

and the CRC Press Web site at

http://www.crcpress.com
To my wife and daughters: Lesley, Elise, and Jessica
— DPK

To Sarah, Sofia, and my parents

— ZIB

To my grandparents: Arno, Harry, Juta, and Maila

— TT

To Valerie
— RV
CONTENTS

Preface

Notation

1 Importing, Summarizing, and Visualizing Data

1.1 Introduction
1.2 Structuring Features According to Type
1.3 Summary Tables
1.4 Summary Statistics
1.5 Visualizing Data
1.5.1 Plotting Qualitative Variables
1.5.2 Plotting Quantitative Variables
1.5.3 Data Visualization in a Bivariate Setting
Exercises

2 Statistical Learning
2.1 Introduction
2.2 Supervised and Unsupervised Learning
2.3 Training and Test Loss
2.4 Tradeoffs in Statistical Learning
2.5 Estimating Risk
2.5.1 In-Sample Risk
2.5.2 Cross-Validation
2.6 Modeling Data
2.7 Multivariate Normal Models
2.8 Normal Linear Models
2.9 Bayesian Learning
Exercises

3 Monte Carlo Methods

3.1 Introduction
Visit https://textbookfull.com
now to explore a rich
collection of eBooks, textbook
and enjoy exciting offers!
 3.2 Monte Carlo Sampling
3.2.1 Generating Random Numbers
3.2.2 Simulating Random Variables
3.2.3 Simulating Random Vectors and Processes
3.2.4 Resampling
3.2.5 Markov Chain Monte Carlo
3.3 Monte Carlo Estimation
3.3.1 Crude Monte Carlo
3.3.2 Bootstrap Method
3.3.3 Variance Reduction
3.4 Monte Carlo for Optimization
3.4.1 Simulated Annealing
3.4.2 Cross-Entropy Method
3.4.3 Splitting for Optimization
3.4.4 Noisy Optimization
Exercises

4 Unsupervised Learning
4.1 Introduction
4.2 Risk and Loss in Unsupervised Learning
4.3 Expectation–Maximization (EM) Algorithm
4.4 Empirical Distribution and Density Estimation
4.5 Clustering via Mixture Models
4.5.1 Mixture Models
4.5.2 EM Algorithm for Mixture Models
4.6 Clustering via Vector Quantization
4.6.1 K-Means
4.6.2 Clustering via Continuous Multiextremal Optimization
4.7 Hierarchical Clustering
4.8 Principal Component Analysis (PCA)
4.8.1 Motivation: Principal Axes of an Ellipsoid
4.8.2 PCA and Singular Value Decomposition (SVD)
Exercises

5 Regression
5.1 Introduction
 5.2 Linear Regression
5.3 Analysis via Linear Models
5.3.1 Parameter Estimation
5.3.2 Model Selection and Prediction
5.3.3 Cross-Validation and Predictive Residual Sum of Squares
5.3.4 In-Sample Risk and Akaike Information Criterion
5.3.5 Categorical Features
5.3.6 Nested Models
5.3.7 Coefficient of Determination
5.4 Inference for Normal Linear Models
5.4.1 Comparing Two Normal Linear Models
5.4.2 Confidence and Prediction Intervals
5.5 Nonlinear Regression Models
5.6 Linear Models in Python
5.6.1 Modeling
5.6.2 Analysis
5.6.3 Analysis of Variance (ANOVA)
5.6.4 Confidence and Prediction Intervals
5.6.5 Model Validation
5.6.6 Variable Selection
5.7 Generalized Linear Models
Exercises

6 Regularization and Kernel Methods

6.1 Introduction
6.2 Regularization
6.3 Reproducing Kernel Hilbert Spaces
6.4 Construction of Reproducing Kernels
6.4.1 Reproducing Kernels via Feature Mapping
6.4.2 Kernels from Characteristic Functions
6.4.3 Reproducing Kernels Using Orthonormal Features
6.4.4 Kernels from Kernels
6.5 Representer Theorem
6.6 Smoothing Cubic Splines
6.7 Gaussian Process Regression
6.8 Kernel PCA
 Exercises

7 Classification
7.1 Introduction
7.2 Classification Metrics
7.3 Classification via Bayes’ Rule
7.4 Linear and Quadratic Discriminant Analysis
7.5 Logistic Regression and Softmax Classification
7.6 K-Nearest Neighbors Classification
7.7 Support Vector Machine
7.8 Classification with Scikit-Learn
Exercises

8 Decision Trees and Ensemble Methods

8.1 Introduction
8.2 Top-Down Construction of Decision Trees
8.2.1 Regional Prediction Functions
8.2.2 Splitting Rules
8.2.3 Termination Criterion
8.2.4 Basic Implementation
8.3 Additional Considerations
8.3.1 Binary Versus Non-Binary Trees
8.3.2 Data Preprocessing
8.3.3 Alternative Splitting Rules
8.3.4 Categorical Variables
8.3.5 Missing Values
8.4 Controlling the Tree Shape
8.4.1 Cost-Complexity Pruning
8.4.2 Advantages and Limitations of Decision Trees
8.5 Bootstrap Aggregation
8.6 Random Forests
8.7 Boosting
Exercises

9 Deep Learning
9.1 Introduction
 9.2 Feed-Forward Neural Networks
9.3 Back-Propagation
9.4 Methods for Training
9.4.1 Steepest Descent
9.4.2 Levenberg–Marquardt Method
9.4.3 Limited-Memory BFGS Method
9.4.4 Adaptive Gradient Methods
9.5 Examples in Python
9.5.1 Simple Polynomial Regression
9.5.2 Image Classification
Exercises

A Linear Algebra and Functional Analysis

A.1 Vector Spaces, Bases, and Matrices
A.2 Inner Product
A.3 Complex Vectors and Matrices
A.4 Orthogonal Projections
A.5 Eigenvalues and Eigenvectors
A.5.1 Left- and Right-Eigenvectors
A.6 Matrix Decompositions
A.6.1 (P)LU Decomposition
A.6.2 Woodbury Identity
A.6.3 Cholesky Decomposition
A.6.4 QR Decomposition and the Gram–Schmidt Procedure .
A.6.5 Singular Value Decomposition
A.6.6 Solving Structured Matrix Equations
A.7 Functional Analysis
A.8 Fourier Transforms
A.8.1 Discrete Fourier Transform
A.8.2 Fast Fourier Transform

B Multivariate Differentiation and Optimization

B.1 Multivariate Differentiation
B.1.1 Taylor Expansion
B.1.2 Chain Rule
B.2 Optimization Theory
 B.2.1 Convexity and Optimization
B.2.2 Lagrangian Method
B.2.3 Duality
B.3 Numerical Root-Finding and Minimization
B.3.1 Newton-Like Methods
B.3.2 Quasi-Newton Methods
B.3.3 Normal Approximation Method
B.3.4 Nonlinear Least Squares
B.4 Constrained Minimization via Penalty Functions

C Probability and Statistics

C.1 Random Experiments and Probability Spaces
C.2 Random Variables and Probability Distributions
C.3 Expectation
C.4 Joint Distributions
C.5 Conditioning and Independence
C.5.1 Conditional Probability
C.5.2 Independence
C.5.3 Expectation and Covariance
C.5.4 Conditional Density and Conditional Expectation
C.6 Functions of Random Variables
C.7 Multivariate Normal Distribution
C.8 Convergence of Random Variables
C.9 Law of Large Numbers and Central Limit Theorem
C.10 Markov Chains
C.11 Statistics
C.12 Estimation
C.12.1 Method of Moments
C.12.2 Maximum Likelihood Method
C.13 Confidence Intervals
C.14 Hypothesis Testing

D Python Primer
D.1 Getting Started
D.2 Python Objects
D.3 Types and Operators
 D.4 Functions and Methods
D.5 Modules
D.6 Flow Control
D.7 Iteration
D.8 Classes
D.9 Files
D.10 NumPy
D.10.1 Creating and Shaping Arrays
D.10.2 Slicing
D.10.3 Array Operations
D.10.4 Random Numbers
D.11 Matplotlib
D.11.1 Creating a Basic Plot
D.12 Pandas
D.12.1 Series and DataFrame
D.12.2 Manipulating Data Frames
D.12.3 Extracting Information
D.12.4 Plotting
D.13 Scikit-learn
D.13.1 Partitioning the Data
D.13.2 Standardization
D.13.3 Fitting and Prediction
D.13.4 Testing the Model
D.14 System Calls, URL Access, and Speed-Up

Bibliography

Index
 PREFACE

In our present world of automation, cloud computing, algorithms, artificial

intelligence, and big data, few topics are as relevant as data science and
machine learning. Their recent popularity lies not only in their applicability
to real-life questions, but also in their natural blending of many different
disciplines, including mathematics, statistics, computer science,
engineering, science, and finance.
To someone starting to learn these topics, the multitude of computational
techniques and mathematical ideas may seem overwhelming. Some may be
satisfied with only learning how to use off-the-shelf recipes to apply to
practical situations. But what if the assumptions of the black-box recipe are
violated? Can we still trust the results? How should the algorithm be
adapted? To be able to truly understand data science and machine learning it
is important to appreciate the underlying mathematics and statistics, as well
as the resulting algorithms.
The purpose of this book is to provide an accessible, yet comprehensive,
account of data science and machine learning. It is intended for anyone
interested in gaining a better understanding of the mathematics and statistics
that underpin the rich variety of ideas and machine learning algorithms in
data science. Our viewpoint is that computer languages come and go, but
the underlying key ideas and algorithms will remain forever and will form
the basis for future developments.
Before we turn to a description of the topics in this book, we would like
to say a few words about its philosophy. This book resulted from various
courses in data science and machine learning at the Universities of
Queensland and New South Wales, Australia. When we taught these
courses, we noticed that students were eager to learn not only how to apply
algorithms but also to understand how these algorithms actually work.
However, many existing textbooks assumed either too much background
knowledge (e.g., measure theory and functional analysis) or too little
(everything is a black box), and the information overload from often
disjointed and contradictory internet sources made it more difficult for
students to gradually build up their knowledge and understanding. We
therefore wanted to write a book about data science and machine learning
that can be read as a linear story, with a substantial “backstory” in the
appendices. The main narrative starts very simply and builds up gradually to
quite an advanced level. The backstory contains all the necessary
background, as well as additional information, from linear algebra and
functional analysis (Appendix A), multivariate differentiation and
optimization (Appendix B), and probability and statistics (Appendix C).
Moreover, to make the abstract ideas come alive, we believe it is important
that the reader sees actual implementations of the algorithms, directly
translated from the theory. After some deliberation we have chosen Python
as our programming language. It is freely available and has been adopted as
the programming language of choice for many practitioners in data science
and machine learning. It has many useful packages for data manipulation
(often ported from R) and has been designed to be easy to program. A
gentle introduction to Python is given in Appendix D.
KEYWORDS

To keep the book manageable in size we had to be selective in our choice

of topics. Important ideas and connections between various concepts are
highlighted keywords via and page references (indicated by a ☞) in the
margin. Key definitions and theorems are highlighted in boxes. Whenever
feasible we provide proofs of theorems. Finally, we place great importance
on notation. It is often the case that once a consistent and concise system of
notation is in place, seemingly difficult ideas suddenly become obvious. We
use different fonts to distinguish between different types of objects. Vectors
are denoted by letters in boldface italics, x, X, and matrices by uppercase
letters in boldface roman font, A, K. We also distinguish between random
vectors and their values by using upper and lower case letters, e.g., X
(random vector) and x (its value or outcome). Sets are usually denoted by
calligraphic letters G , H . The symbols for probability and expectation are
ℙ and E , respectively. Distributions are indicated by sans serif font, as in
Bin and Gamma; exceptions are the ubiquitous notations N and U for the
normal and uniform distributions. A summary of the most important
symbols and abbreviations is given on Pages xvii–xxi.
☞ xvii
Data science provides the language and techniques necessary for
understanding and dealing with data. It involves the design, collection,
analysis, and interpretation of numerical data, with the aim of extracting
patterns and other useful information. Machine learning, which is closely
related to data science, deals with the design of algorithms and computer
resources to learn from data. The organization of the book follows roughly
the typical steps in a data science project: Gathering data to gain
information about a research question; cleaning, summarization, and
visualization of the data; modeling and analysis of the data; translating
decisions about the model into decisions and predictions about the research
question. As this is a mathematics and statistics oriented book, most
emphasis will be on modeling and analysis.
We start in Chapter 1 with the reading, structuring, summarization, and
visualization of data using the data manipulation package pandas in Python.
Although the material covered in this chapter requires no mathematical
knowledge, it forms an obvious starting point for data science: to better
understand the nature of the available data. In Chapter 2, we introduce the
main ingredients of statistical learning. We distinguish between supervised
and unsupervised learning techniques, and discuss how we can assess the
predictive performance of (un)supervised learning methods. An important
part of statistical learning is the modeling of data. We introduce various
useful models in data science including linear, multivariate Gaussian, and
Bayesian models. Many algorithms in machine learning and data science
make use of Monte Carlo techniques, which is the topic of Chapter 3.
Monte Carlo can be used for simulation, estimation, and optimization.
Chapter 4 is concerned with unsupervised learning, where we discuss
techniques such as density estimation, clustering, and principal component
analysis. We then turn our attention to supervised learning in Chapter 5, and
explain the ideas behind a broad class of regression models. Therein, we
also describe how Python’s statsmodels package can be used to define and
analyze linear models. Chapter 6 builds upon the previous regression
chapter by developing the powerful concepts of kernel methods and
regularization, which allow the fundamental ideas of Chapter 5 to be
expanded in an elegant way, using the theory of reproducing kernel Hilbert
spaces. In Chapter 7, we proceed with the classification task, which also
belongs to the supervised learning framework, and consider various
methods for classification, including Bayes classification, linear and
quadratic discriminant analysis, K-nearest neighbors, and support vector
machines. In Chapter 8 we consider versatile methods for regression and
classification that make use of tree structures. Finally, in Chapter 9, we
consider the workings of neural networks and deep learning, and show that
these learning algorithms have a simple mathematical interpretation. An
extensive range of exercises is provided at the end of each chapter.

Python code and data sets for each chapter can be downloaded from the
GitHub site: https://github.com/DSML-book

Acknowledgments
Some of the Python code for Chapters 1 and 5 was adapted from [73]. We
thank Benoit Liquet for making this available, and Lauren Jones for
translating the R code into Python.
We thank all who through their comments, feedback, and suggestions
have contributed to this book, including Qibin Duan, Luke Taylor, Rémi
Mouzayek, Harry Goodman, Bryce Stansfield, Ryan Tongs, Dillon Steyl,
Bill Rudd, Nan Ye, Christian Hirsch, Chris van der Heide, Sarat Moka,
Aapeli Vuorinen, Joshua Ross, Giang Nguyen, and the anonymous referees.
David Grubbs deserves a special accollade for his professionalism and
attention to detail in his role as Editor for this book.
The book was test-run during the 2019 Summer School of the Australian
Mathematical Sciences Institute. More than 80 bright upper-undergraduate
(Honours) students used the book for the course Mathematical Methods for
Visit https://textbookfull.com
now to explore a rich
collection of eBooks, textbook
and enjoy exciting offers!
Machine Learning, taught by Zdravko Botev. We are grateful for the
valuable feedback that they provided.
Our special thanks go out to Robert Salomone, Liam Berry, Robin
Carrick, and Sam Daley, who commented in great detail on earlier versions
of the entire book and wrote and improved our Python code. Their
enthusiasm, perceptiveness, and kind assistance have been invaluable.
Of course, none of this work would have been possible without the loving
support, patience, and encouragement from our families, and we thank them
with all our hearts.
This book was financially supported by the Australian Research Council
Centre of Excellence for Mathematical & Statistical Frontiers, under grant
number CE140100049.

Dirk Kroese, Zdravko Botev,

Thomas Taimre, and Radislav Vaisman
Brisbane and Sydney
 NOTATION

We could, of course, use any notation we want; do not laugh at

notations; invent them, they are powerful. In fact, mathematics
is, to a large extent, invention of better notations.
Richard P. Feynman

We have tried to use a notation system that is, in order of importance,

simple, descriptive, consistent, and compatible with historical choices.
Achieving all of these goals all of the time would be impossible, but we
hope that our notation helps to quickly recognize the type or “flavor” of
certain mathematical objects (vectors, matrices, random vectors, probability
measures, etc.) and clarify intricate ideas.
We make use of various typographical aids, and it will be beneficial for
the reader to be aware of some of these.

• Boldface font is used to indicate composite objects, such as column

vectors x = [x1,…, xn]⊤ and matrices X = [xij]. Note also the difference
between the upright bold font for matrices and the slanted bold font for
vectors.

• Random variables are generally specified with upper case roman letters
X, Y, Z and their outcomes with lower case letters x, y, z. Random
vectors are thus denoted in upper case slanted bold font: X = [X1,…,
X ]⊤.
n
• Sets of vectors are generally written in calligraphic font, such as χ, but
the set of real numbers uses the common blackboard bold font ℝ.
 Expectation and probability also use the latter font.

• Probability distributions use a sans serif font, such as Bin and Gamma.
Exceptions to this rule are the “standard” notations N and U for the
normal and uniform distributions.

• We often omit brackets when it is clear what the argument is of a

function or operator. For example, we prefer EX to E [X ].
2 2

• We employ color to emphasize that certain words refer to a dataset,

function, or package in Python. All code is written in typewriter font.
To be compatible with past notation choices, we introduced a special
blue symbol X for the model (design) matrix of a linear model.

• Important notation such as T g, g* is often defined in a mnemonic

,
way, such as T for “training”, g for “guess”, g* for the “star” (that is,
optimal) guess, and ℓ for “loss”.

• We will occasionally use a Bayesian notation convention in which the

same symbol is used to denote different (conditional) probability
densities. In particular, instead of writing fX(x) and fX | Y(x | y) for the
probability density function (pdf) of X and the conditional pdf of X
given Y, we simply write f (x) and f (x | y). This particular style of
notation can be of great descriptive value, despite its apparent
ambiguity.

General font/notation rules

x scalar
x vector
X random vector
X matrix
χ set
x̂ estimate or approximation
x* optimal
x
¯ average

Common mathematical symbols

∀ for all
∃ there exists
∝ is proportional to
⊥ is perpendicular to
~ is distributed as
iid ,~ are independent and identically distributed as
˜
iid
approx. is approximately distributed as
~

∇f gradient of f
∇ 2f Hessian of f
f ∈ Cp f has continuous derivatives of order p
≈ is approximately
≃ is asymptotically
≪ is much smaller than
⊕ direct sum
⊙ elementwise product
∩ intersection
∪ union
≔,=: is defined as
a.s.
converges almost surely to
−
→
d
converges in distribution to
→
P
converges in probability to
→
Lp
converges in Lp-norm to
→

║·║ Euclidean norm

⌈ x⌉ smallest integer larger than x
⎿ x⏌ largest integer smaller than x
x+ max{ x, 0}

Matrix/vector notation
 A⊤,x⊤ transpose of matrix A or vector x
A-1 inverse of matrix A
A+ pseudo-inverse of matrix A
A-⊤ inverse of matrix A⊤ or transpose of A-1
A≻0 matrix A is positive definite
A ≥ 0 matrix A is positive semidefinite
dim( x) dimension of vector x
det(A) determinant of matrix A
|A| absolute value of the determinant of matrix A
tr(A) trace of matrix A

Reserved letters and words

ℂ set of complex numbers
d differential symbol
E expectation
e the number 2.71828 …
f probability density (discrete or continuous)
g prediction function
𝟙{A} or 𝟙 indicator function of set A
A
i the square root of –1
ℓ risk: expected loss
Loss loss function
ln (natural) logarithm
ℕ set of natural numbers {0,1,…}
O big-O order symbol: f (x) = O (g (x)) if | f (x)| ⩽ αg(x) for
some constant α as x → a
o little-o order symbol: f (x) = o(g(x)) if f (x)/g(x) → 0 as x → a
ℙ probability measure
π the number 3.14159 …
ℝ set of real numbers (one-dimensional Euclidean space)
ℝn n-dimensional Euclidean space
ℝ+ positive real line: [0, ∞)
Random documents with unrelated
content Scribd suggests to you:
“You seem a little old for the part,” she suggested.
A young man of a rather assertive Bohemian appearance stood
before her.
“Hello!” was his greeting. The tone denoted surprise, rather than
familiarity. He hastily added: “Excuse me—is Victor—Mr. Baron—in?”
Bonnie May perceived that he was not quite comfortable, not at all
self-possessed. He seemed to her a strange person to be calling on
any of the Barons. Still, he seemed rather human.
“I’ll see,” she said. “Please step inside.” She would make him wait in
the hall, she decided.
“Tell him, please, that Baggot has called—that I’ve brought the first
act of my play.”
“A play! Oh!”
Again she hurried up the stairs; this time with unconcealed
eagerness. When she entered Mrs. Baron’s room she hesitated. “If
you’ll excuse me—” she faltered. “I’m looking for Mr. Victor.”
Mrs. Baron sat more erect, the open volume in her lap. “Forming a
little organization down-stairs?” she asked.
“Some one’s called for Mr. Victor. I wanted to tell him.”
“Very well. He’s in the library.” She nodded toward the adjoining
room.
Victor was alone in the library. He was in the attitude of one who is
about to write, but he was not writing. He was glowering at the paper
before him.
He sprang to his feet eagerly when Bonnie May appeared.
“I’ve been thinking about you,” he said. “Flora has, too. We meant to
come and find you before long and get you away from Mrs. Shepard.
We didn’t want to seem too eager, you know. We wanted to wait until
the governess——”
Bonnie May did not wait for him to finish; indeed, he seemed to have
difficulty about finishing. “Mr. Baggot has called,” she said. “It’s about
a play.” She was breathing uneasily. “And couldn’t I sit with you and
listen, please?” she added.
“Oh! Baggot! Baggot is one of my crosses, Bonnie May. Couldn’t you
shut the door in his face? It would be quite proper. He is one of those
silly fellows who think they are destined to write great plays. Couldn’t
you go down and put him out?”
She looked at him steadily without a word. She was smiling a little
scornfully.
“Very well. Suppose you go and ask him to come up—this time.”
“And—do let me come too! They’ve often let me listen when new
plays were being read.”
“Such wanton cruelty!” He shook his head slowly, as if it were quite
incredible. “Oh, well you may come, too,” he added.
Mrs. Baron glanced up from her book again when Bonnie May and
Baggot passed through the room. She spoke to Baggot in the most
casual manner. Bonnie May concluded that he must be a somewhat
frequent visitor. Mrs. Baron was quite frank in her indifference to him.
“I think you’ll find Victor in the library,” she said. She glanced
pointedly at the manuscript in his hand and frowned. “And would you
mind closing the door when you go in?”
Mrs. Baron achieved her cruelties sometimes with such a naïve
directness that they seemed to many people like a kind of high
breeding.
Baggot stepped gingerly into the next room, followed eagerly by
Bonnie May. He was closing the door softly when Baron greeted him.
“Hello, Baggot. Done something great again, of course?”
“Yes, I have!” retorted Baggot angrily. He wouldn’t endure Baron’s
bad manners, no matter how he might receive the bad manners of
Baron’s mother. “You’re going to say so, too. I’ve got the first act
finished. I’ve only got to fill in the scenario of the other acts, and I’ve
got the greatest play that ever came out of America.”
Baron smiled wearily. “And I’m to listen while you read the first act of
the greatest play, etc.?”
“Yes—and you’re to agree with me, too. I don’t see anything great in
your sneering at me all the time!” He pulled up a chair and sat down
so that his knees almost touched Baron’s.
Obviously, they were a pair of young men on very intimate terms.
Bonnie May slipped into a remote corner of the room and climbed
into a big chair. Her hand supported her chin; her eyes were
luminous. She did not mean to miss a word.
And Baggot began to read. His face was almost tortured with
nervous energy. He handled the pages as if they were in hopeless
confusion, yet he brought order out of them swiftly.
The reading proceeded ten minutes, twenty minutes, half an hour.
Baggot read with profound confidence and belief. His staccato tones
fairly hurled the words of the play at his auditors. Baron had put
away his cynic attitude. He had become deeply impressed. He had
even forgotten that it was his favorite pose not to seem deeply
impressed by anything.
Bonnie May was like one in a beautiful dream. She was not only
listening to the play; she was living it.
And then her dream was broken in a manner which filled her mind
with almost blank astonishment.
Mrs. Baron appeared in the doorway.
“Bonnie May,” she announced, “I think it’s high time for a little girl to
be in bed.”
 CHAPTER VI
CONCERNING A FROCK

It might have been, and should have been, apparent to the several
members of the Baron household that Bonnie May had been giving
an admirable exhibition of self-repression from the moment she had
entered the house.
A change came at last—when Mrs. Baron disturbed the reading of
the play and announced, at nine o’clock, that it was “high time for a
little girl to be in bed.”
Mrs. Baron couldn’t possibly have realized how Bonnie May had
been accustomed to divide her hours between sleeping and waking.
The guest had spent her life among player people, whose active
hours begin at noon or later, and who do not deem the day ended
until after midnight—sometimes far later than midnight. Nor had it
been found convenient—or needful—by Bonnie May’s fellow workers
to make any exception to the rule on her behalf. She had been one
of them, and she had fared well and pleasantly.
Thus it was that when Mrs. Baron appeared, somewhat like a bolt
out of a clear sky, the child gave way to overwhelming rebellion.
“I’m not used to going to bed at this hour,” she declared bluntly. She
arose and stood by her chair, like a soldier by his guns, as the saying
is. And taking in the inexorable expression in Mrs. Baron’s eyes, she
turned appealingly to Baron. She was relying upon him to help her.
“Couldn’t she—” began Baron weakly, and added, quite without
conviction: “You know it’s Saturday night, mother!” He was glad he
had thought of its being Saturday, though he couldn’t see why that
should make very much difference. He really believed his mother’s
position was strong enough, if she had only gone about the matter
more tactfully.
“Saturday night doesn’t make any difference,” declared Bonnie May,
her rebellion now including Baron in its scope. “It just isn’t a
reasonable bedtime.”
Baron felt ready to surrender. “Anyway, it won’t be so bad just for
one night,” he ventured.
“Never mind, Victor,” said Mrs. Baron pointedly. She addressed
herself to Bonnie May. “What you’ve been accustomed to may not be
quite so important as what you ought to become accustomed to,”
she said. “Come!”
The child sauntered thoughtfully from the room. She had been
impressed by the fact that even Baron had not seemed surprised by
the suggestion that she ought to go to bed. She was trying to
comprehend the situation. After all, people who were not of the
profession had ways of their own, she realized. If they had all
decided to go to bed, she wouldn’t have minded so much. But they
were laying down a special law for her.
Rebellion triumphed again. In Mrs. Baron’s room she halted. “Where
am I to sleep?” she inquired.
“I think you heard me tell Mrs. Shepard to prepare a room.”
“In the attic? Yes. But I’m not going to sleep there.”
“Indeed, you are.”
“I beg your pardon! Not under any circumstances!”
Mrs. Baron lifted her fingers to her lips and coughed—a very inexpert
cough. “You’ll have to do as I tell you, you know.” She resumed a
resolute march toward the hall, her hand pressed firmly against
Bonnie May’s back.
The child jerked away with a sense of outrage. She had never been
treated so before.
“Truly, you’ll have to obey me,” repeated Mrs. Baron.
Bonnie May was alarmed; she quite lost control of herself. “Stop your
kiddin’!” she said with a catch in her voice. She tried to say it
playfully, but her self-possession was gone. Her remark had
sounded simply offensive, indelicate.
“And I can’t permit you to use such language, either!” declared Mrs.
Baron.
The dismayed guest pressed her hands to her eyes as if she were
trying to think clearly.
Then she made a rush for the stairway!
Mrs. Baron put dignity aside long enough to pursue her, to seize her
by the arm. She was becoming outraged, greatly indignant. “What do
you mean to do?” she demanded, her voice trembling slightly.
“I’m quitting.”
“You’re——”
“I won’t stay here!”
The distressed old gentlewoman tried to calm herself. “Where do you
think of going?” she asked.
“Anywhere—to the theatres. Any company in town will be glad to
have me. They will know who I am. They—they are the kind of
people who will appreciate me!” The words were spoken in a tone of
heart-break, of despair.
Mrs. Baron afterward confessed to members of her family that for the
first time in her life she felt completely helpless. She was, in truth, a
somewhat childish person in many ways, and she was not
accustomed to any unpleasantnesses save those which she created
for others.
At any rate, she swallowed with difficulty—and surrendered. “It’s a
very small point, after all,” she said ungraciously. “Go into my room.
Flora will look after you.” She spoke coldly, all her interest seemingly
withdrawn.
And just as the guest disappeared into Mrs. Baron’s sitting-room,
Flora came almost stealthily up the stairs.
“I wish you’d put that little limb of Satan to bed,” she said. Flora saw
that her mother’s hand, on the balustrade, trembled.
“Where shall I put her?” she inquired.
“Anywhere! just so you get her covered up for the night.”
Flora paused, her eyes uneasily seeking her mother’s.
“I’m afraid you’re angry with me, mother,” she said humbly.
“With you? Certainly not.”
Flora was puzzled. Her mother had long ago declared that Mr. Addis
must not be accepted as a visitor. Did she know that he had just
gone? She was about to enter her mother’s sitting-room when
something prompted her to turn.
“You knew Mr. Addis called, didn’t you?” she asked.
Mrs. Baron’s face flamed again. “Knew it? Certainly, I didn’t know it!
I’ve told Mrs. Shepard—I don’t intend that he shall annoy you!”
“Oh, mother! He doesn’t! And I think Mrs. Shepard didn’t know, this
time. Bonnie May went to the door and let him in. She called me
down-stairs without telling me who it was.” Flora surveyed her
mother yearningly, yet with a kind of gentle courage. “I don’t believe
in hiding things from you, mother. But I was glad to see him.”
Mrs. Baron looked grimly toward her own door. “She let him in! Very
well. Put her to bed!”
She descended the stairs with dignity. She must have been thinking
of future victories rather than of past defeats.

When Flora entered the sitting-room she found Bonnie May standing
in uneasy contemplation.
“Mother says I’m to put you to bed,” said Miss Baron.
“Why didn’t she go ahead and put me to bed herself?”
Flora perceived that the question was not wanting in sincerity. She
decided to answer quite honestly.
“I think,” she ventured gently, “you must have said something to vex
her.”
“Not at all. She tried to vex me. I behaved very properly.”
Flora sighed and shook her head slowly; but she was smiling, too.
She was wondering what it really was that had gone wrong.
“Possibly you didn’t want to obey her?” she ventured.
The child’s brow puckered. “But why should I want to obey her?”
“Why—because she’s going to be good to you, I’m sure.”
“Well, I mean to be good to her, too—if she’ll let me. And I don’t ask
her to obey me.”
“But it’s different. She’s an old lady.”
“Well, I’ve got no patience with old people. It’s all right, just as a part,
but there’s no use putting it on all the time.”
“But, dear,” implored Flora, drawing the child within the curve of her
arm, “don’t say that! I know you mean to be nice and kind, but truly
you don’t understand. We must all grow old some time—even you
will get to be old.”
The guest gave deliberate thought to this; then her expression
became resolute. “Well, if they ever hang any gray hairs on me
they’ll have to catch me when I’m asleep—I’ll tell you that right now.”
Miss Baron was not encouraged to argue the point any further. She
resumed the subject of going to bed.
“You know I’m to have his room—your brother’s?” the guest insisted.
“Mother said you might sleep where you liked.”
“Did she say that?”
“Almost exactly.”
“Well, where is that attic room?”
“It’s up one more flight of stairs—under the roof.”
The child looked quite wistful and earnest, and then her words came
with conviction. “I just couldn’t sleep up there. Attics are where
misers sleep, and poor children. It’s where people die of hunger and
cold. It’s never the right kind of people. Come, let’s go to his room.”
And so they did.
“You won’t mind my helping you?” pleaded Flora.
“Helping me?”
“To undress, you know—and to be tucked in!”
The guest looked at her unresponsively. “But I’ve been used to doing
that for myself,” she said.
Flora quickly stooped and took her into her arms impulsively. “Dear
child,” she cried, her voice tremulous, “let me do it to-night! I think
you’ll love it—and I’ll love it, too.” She drew the perplexed face
almost roughly against her own.
She did not wait to be refused. She hurried into the bathroom and
busied herself; she was singing a little crooning song. There was
also the noise of water splashing into the tub.
She reappeared presently. “The water is ready—for your bath, you
know, and I’ve left one of my nighties there for you.” She smiled
happily. “Of course it will be too big. I’ll make you some little ones
soon.”
The seeming perversion of the child asserted itself again. “I usually
take my bath in the morning,” she said a little stiffly; but she saw how
the glad light in Miss Baron’s eyes wavered, and she added quickly,
“but it will be all right.” And she went out into the bathroom.
When she reappeared after a rather long time she was smiling
radiantly. She had on Flora’s nightgown, soft and white, with pink
ribbons. She held it daintily up before her feet, and glanced back at
the train that dragged behind. “Isn’t it lovely!” she said.
“It is, dear,” said Flora.
She had turned the white coverlet and the sheet down. Now she
watched the child scramble up into the bed. She wanted to help, but
she refrained.
“Would you like me to tell you a story?” asked Flora.
Bonnie May looked at her swiftly, incredulously. “No!” she said. She
burst out into riotous laughter. “I’m not an infant,” she explained.
Flora flushed. “Very well,” she said gently. Yet she lingered in the
room a little while. She put some of Victor’s masculine decorations
out of sight. She adjusted the blind. She was about to extinguish the
light when she looked again at the strange guest.
The child’s eyes were fixed upon her widely, wonderingly.
“You lovely thing!” said Bonnie May.
“Good night, dear!” said Flora. And then she knew that the child
wished to speak to her, and she went over and bent above the bed.
“What is it, Bonnie May?” she asked.
The child stared before her in silence for a moment and then the
words came. “I wished so much that she would love me!” she said. “I
tried so hard....”
Flora slipped her hand under the guest’s head. “I’ll tell you a secret,”
she whispered. “If she hadn’t cared for you, she would have been
quite polite; she would have been wonderfully gracious. She was
ungracious and unkind because—because she loved you, dear. It
seems absurd, doesn’t it? But I know.”

It was an absurd theory, perhaps; yet there was certainly needed

some explanation of Mrs. Baron’s course later in the evening.
The house became quiet after a time. The rumbling voices in the
library ceased and Baggot, with meticulous circumspection, wended
his way down the stairs and was gone. Later, Victor emerged from
the library and disappeared for the night. Baron, Sr., came in and sat
and smoked awhile—and retired. Flora sat in the sitting-room
lingeringly, gazing pensively at a book without turning its pages, and
at length she arose and kissed her mother’s cheek and said good
night.
And then Mrs. Baron tiptoed into another room and rummaged in a
bureau drawer and found a gay piece of gingham which had been
waiting its time to be useful. With this in her hands she returned to
her sitting-room, and spread work materials upon her table. And with
patience and fortitude and a kind of rapt self-absorption she worked
far into the night.

The usual Sunday-morning quietude of the mansion was disturbed

somewhat when the family again assembled. An extraordinary event
had occurred.
Mrs. Baron had sat up late the night before and had made a Dress.
In announcing the fact she had pronounced the word in such a
manner that the use of the capital letter is fully justified. She
displayed the Dress for the admiration of her son and her daughter,
and her husband. And finally she generously relinquished it to Flora.
“You may give it to her,” she said rather loftily.
Bonnie May had not yet appeared.
Flora, knocked softly on the guest’s door and without waiting went
into the room, displaying the new garment rather conspicuously.
“What’s that?” inquired Bonnie May dubiously.
“It’s a new dress for you.”
“It was never made for me,” affirmed the child with conviction.
“Indeed, it was. Mother sat up ever so late last night and made it for
you.”
“Well, that, of course, was a matter I should have been consulted
about.”
Bonnie May was now sitting on the edge of the bed, trying to make
the toes of one foot come in contact with the floor. Miss Baron sat on
a low chair in the middle of the room, the new dress spread across
her knees.
“But you’re glad, aren’t you?” she asked.
“I’m glad in a way. I’m glad that anybody so disagreeable could really
try to do you a good turn.” Clearly, each day was a new day, with
Bonnie May.
“But, dear child, mother won’t seem disagreeable to you when you
come to know her. It hurts me to have you speak so of her—truly it
does. And I think she must have worked until she was very weary,
making the dress for you.”
“I appreciate all that,” the guest hastened to explain, genuine
compunction in her voice. “But you see, the dress isn’t at all
suitable.”
“I’m sure you’ll like it much better when you try it on.”
“Take my word for it—it won’t do.”
Miss Baron felt for the moment as if she could have pounced upon
the child and spanked her. But she noticed how one curl fell outside
her ear, and how the eyes and voice were profoundly earnest, and
how the attitude was eloquent of a kind of repentance before the
fact.
And so she said: “Won’t you do something for me that will please me
better than anything else I can think of—something that will take only
a minute?”
Bonnie May looked at her meditatively—and then began to laugh
quite riotously! “You don’t look the part!” she gurgled in justification.
“What part, please?” The question was put somewhat blankly.
“You’re talking like a—oh, a Lady Clare, and you haven’t even got
your shoes buttoned up!”
Miss Baron slowly regarded her shoes; then her glance travelled
calmly to Bonnie May; then she rather dully inspected the dress that
lay across her knees. Her countenance had become inscrutable.
She turned away from the guest’s scrutiny, and after a moment she
arose slowly and left the room, carrying the dress with her.
She did not stop to define her feelings. She was wounded, but she
felt sharp resentment, and she was thinking rebelliously that she was
in no degree responsible for Bonnie May. Still ... her sense of justice
stayed her. She had the conviction that the child’s remark, if
inexcusably frank, was a fair one. And it had been made so joyously!
However, she meant to go to her mother with a request to be
excused from any further humiliation as Bonnie May’s handmaiden.
But before she had proceeded half a dozen steps she began to fear
even greater disaster, if Mrs. Baron should undertake to be the
bearer of the rejected dress.
It would be a victory worth working for, if she could overcome the
fastidious guest’s prejudice.
She went to her room and carefully buttoned her shoes and made
other improvements in her toilet. Then she went back into Bonnie
May’s presence.
“I was untidy,” she confessed. “I hope you’ll excuse me.” She was
smoothing out the new dress. “You see, I only meant to wear my
every-day shoes until after breakfast, and then put on my good
shoes, for Sunday-school and church. And I’ve been very busy.”
Bonnie May pondered this judicially. “It’s lovely of you to be so nice
about it,” she finally admitted, “but I’m afraid I don’t get your idea....”
She frowned. “Every-day shoes’ and ‘Sunday shoes,’” she repeated
vaguely.
“Well?” said Flora persuasively.
“Don’t you like to be as good on Saturday as on Sunday?”
“Why, yes—just as good, certainly.” Flora was looking bewildered.
“And on Friday, and on other days?”
“Yes, I think so.”
“Well, why shouldn’t you wear your ‘good’ shoes all the week, then?”
“But people must look nicer on Sundays than on other days.”
“I don’t see why. If you only look nice, I don’t see what’s the good.
And if you really are nice, I think the nice shoes might help all the
time.”
“What I mean is,” persisted Flora patiently, “I don’t like to work in my
nice shoes.” She brought this out somewhat triumphantly.
“That’s funny. That’s the very time I like to look my best. Nothing is
as important as your work, is it?”
Flora was almost in despair. “I doubt if I ever thought of it in just that
light,” she admitted. “I’ll think it over, if you’ll try the dress on—and if
you don’t like it, off it comes!”
“Well, all right.” (This with a sudden calm which was not reassuring.)
Flora slipped the gingham dress into place, and patted it here and
there with the air of one who admires, and viewed it with her head
inclined a little, as women do in such a situation. “It’s the dearest
thing!” she said honestly. “Now come and see how you look.”
The mirror was a little high. She lifted Bonnie May to a chair.
She was alarmed by what ensued. The child stared fixedly, with
incredulous eyes in which a great horror grew.
“Oh, Lord!” she cried, clapping her hands over her eyes. “Take it off!
Take it off!”
“What in the world is the matter?” demanded Flora.
“She asks me what is the matter! Oh, heavens!” Bonnie May jumped
down from the chair and turned her back to the mirror. She was
wringing her hands.
“I don’t understand at all!” exclaimed Miss Baron hopelessly.
“You might!” was the emphatic rejoinder. “Do you suppose I want to
play that kind of a part—here? It might do for the little sister of a
sewing-machine girl, or a mountain-pink with her hair in knobs. But it
wouldn’t do for anything else. If you was only one of the populace, a
costume like that would cause a scream! If you don’t understand it,
take my word for it. I can’t wear it! I ask you to take it off!”
Miss Baron became very quiet. She became thoughtful, too. She had
not failed to catch the drift of these exaggerated words. There was
something prim, something rudimentary, about the dress. Color
suffused her cheeks; she hung her head. She felt a forlorn inclination
to laugh. From a vantage point behind the child she began to remove
the gingham dress.
It was inappropriate. She had to admit it. It was a dress for a
Gretchen; for the Cinderella of the kitchen, rather than the princess
of the coach-and-four. It wasn’t becoming at all.
 CHAPTER VII
A SUNDAY MORNING

The Barons were the kind of family that have just one morning
newspaper left at their door on Sunday, and who believe that it
contains everything that ought to concern them in any way—that
whatever is published in any other newspaper is to be regarded with
scepticism, or lightly discredited.
Yet on this particular Sunday morning Victor Baron arose early and
intercepted the paper-carrier, and amazed that industrious youth by
buying a copy of every journal he carried.
With this not inconsiderable burden under his arm he betook himself
to the library and began an eager search for certain information.
He scanned all the advertising columns systematically, and then
turned to the news departments.
A great heap of discarded “sections” grew about him as he
progressed, and little by little a look of troubled anticipation vanished
from his eyes. The last section of the last paper was cast away with
an air of triumph.
He hadn’t been able to find a single word about any child who was
lost, or who had strayed, or who had been stolen!
“Good!” he exclaimed, and he looked with great relief at the heap of
papers about him, their splotches of color and assertive head-lines
having no further interest for him. He smiled complacently.
In the meantime, in the sunny sitting-room up-stairs, Flora had
broken the news to Mrs. Baron—the news touching Bonnie May and
the new dress.
It had been a very difficult thing to do, because Mrs. Baron was
always at her worst on Sunday mornings.
It was on Sunday mornings that she felt most keenly the lapse of the
neighborhood from former glories to a condition of sordid griminess.
It was on these mornings that she fared forth to the old church, only
three blocks away, in which the best people in town had formerly
worshipped, but which had been deserted by nearly all the old
parishioners.
It was Mrs. Baron’s contention that it was indelicate, to say the least,
for people to desert a church. There were things in the church life,
she maintained, which could not be transplanted, and which
constituted the very warp and woof of the domestic as well as the
social foundations. She had come to regard herself as a kind of
standard-bearer in this relationship, and she attended services
somewhat ostentatiously, with the belief that she was not only
lending her influence, but administering a rebuke as well. Ignoring
the protests of her family, she had even consented to play the organ
for the Sunday-school services. As a young lady she had learned to
read music, as a matter of course, and though she possessed no
musical intelligence, and had found it impossible to regain the old
manual skill she had once possessed, she played the simple hymns
with a kind of proud rigor, because she believed her participation in
the services in this direction must impart an authority to the
proceedings which the abler playing of some obscure individual
could not have imparted.
Indeed, Mrs. Baron was a personage on Sunday mornings; a gallant
general leading a forlorn hope proudly and firmly.
When Flora confessed to her that the dress had been rejected, she
was too greatly amazed to say a great deal. She had also entered
upon her stoic mood—her Sunday-morning mood.
“You see, she is simply determined not to get along,” she declared
with finality. She took the dress into her own hands and regarded it
critically. “Do you see how carefully the feather-stitching is done?”
she demanded.
“Yes,” agreed Flora, “the—the feather-stitching is beautiful. But
really, I don’t believe she is simply perverse. If you could have seen
the dismay in her eyes—” Flora smiled at the recollection.
“I’ve seen women like that,” Mrs. Baron continued, “women who like
to make difficulties; who go into hysterics over little things. It’s always
just a lack of sense—that’s all it is.”
“Yes—or temperament. I expect there’s a good deal in what people
call temperament. I didn’t know children had it so much, but Bonnie
May isn’t like other children. Maybe she has a good deal of
temperament.”
They examined the dress together without any very definite purpose.
“She ought to know she can’t go on wearing that silly thing she came
here in,” was Mrs. Baron’s next comment.
“She must realize that,” agreed Flora. She added casually: “I think
something soft, with a little color in it, might please her. You might let
me try next time.”
This was the wrong note again. “As if I weren’t capable of making a
child’s dress!” protested Mrs. Baron.
“I only meant it would be fair to divide the work,” Flora explained
gently. “I didn’t mean I could do it better.”
As if her anger had been effectually checked in that direction, Mrs.
Baron hit upon another possible grievance. “And she’s going to
Sunday-school to-day,” she affirmed in a tone which seemed to take
account of difficulties. “We’ve done our best to dress her decently.
And I don’t intend to humor a little pagan as long as she’s in a
Christian household.”
“But in that—that peculiar dress?” faltered Flora. She had a vision of
Bonnie May in her fantastic old frock associating with the prim
children of poverty who were now the mainstay of the Sunday-
school.
“She may walk with Mrs. Shepard. People may believe she belongs
to her, if they want to.”
“Oh, mother!” There was something almost despairing in Flora’s
tone.
“It’s the best we can do. I mean to do my duty—and I’m not willing to
look ridiculous.”
Again Miss Baron perceived breakers ahead. If the child conceived
the idea that she was being commanded to go anywhere she would
very probably develop new methods of resistance. If she were
politely invited to accompany other members of the household to
church, she might decide to be altogether gracious.
She entertained a lingering regret that the guest could not be
persuaded to wear the new dress—in which, certainly, she would be
conspicuous enough, but not quite in a flaunting fashion. She even
thought of Victor, and wondered if he might not be able to prevail
upon the child to accede to the wishes of her elders. But upon
second thought she decided not to involve her brother in a phase of
the problem which did not touch him. She suspected there would be
other phases, more in his line, in due time.

In the meanwhile, the object of all this solicitous thought was

leisurely preparing to make her appearance.
That she had no fresh raiment to put on was not particularly
disquieting. The fact that it was a Sunday morning made no
difference to her at all. Certainly she needed fresh linen, but this, she
philosophically concluded, would be provided within another day or
two. Her shoes were quite new and neat, and she was by no means
ashamed of the dress which now constituted her complete wardrobe.
On a chair by her bed she made discoveries. There was a fresh
towel; a little package which obviously contained a tooth-brush; a
box of tooth-powder, and—crowning gift—a new hair-ribbon of
adorable width and hue.
She tucked these things under one arm, and with her free hand she
carefully gathered Flora’s long nightgown away from her feet. Then
she started to the bathroom.
In the hall she paused to be sure that the way was clear.