(eBook PDF) Business Analytics Using SAS

Enterprise Guide and SAS Enterprise Miner pdf

download

https://ebooksecure.com/product/ebook-pdf-business-analytics-
using-sas-enterprise-guide-and-sas-enterprise-miner/

Download more ebook from https://ebooksecure.com

We believe these products will be a great fit for you. Click
the link to download now, or visit ebooksecure.com
to discover even more!

(eBook PDF) SAS Certified Specialist Prep Guide: Base

Programming Using SAS 9.4

http://ebooksecure.com/product/ebook-pdf-sas-certified-
specialist-prep-guide-base-programming-using-sas-9-4/

(Original PDF) SAS Certification Prep Guide Base

Programming for SAS9, Fourth Edition

http://ebooksecure.com/product/original-pdf-sas-certification-
prep-guide-base-programming-for-sas9-fourth-edition/

(eBook PDF) (AUCM) Enterprise, Innovation and Markets

3rd

http://ebooksecure.com/product/ebook-pdf-aucm-enterprise-
innovation-and-markets-3rd/

(eBook PDF) (AUCM) Enterprise, Innovation and Markets

4th Edition

http://ebooksecure.com/product/ebook-pdf-aucm-enterprise-
innovation-and-markets-4th-edition/
(eBook PDF) Concepts in Enterprise Resource Planning
4th Edition

http://ebooksecure.com/product/ebook-pdf-concepts-in-enterprise-
resource-planning-4th-edition/

(eBook PDF) Technology Ventures: From Idea to

Enterprise 5th Edition

http://ebooksecure.com/product/ebook-pdf-technology-ventures-
from-idea-to-enterprise-5th-edition/

People-centric security : transforming your enterprise

security culture 1st Edition Hayden - eBook PDF

https://ebooksecure.com/download/people-centric-security-
transforming-your-enterprise-security-culture-ebook-pdf/

(eBook PDF) Business Statistics in Practice Using Data,

Modeling Analytics 8th

http://ebooksecure.com/product/ebook-pdf-business-statistics-in-
practice-using-data-modeling-analytics-8th/

(eBook PDF) Using Excel for Business and Financial

Modelling: A Practical Guide 3rd Edition

http://ebooksecure.com/product/ebook-pdf-using-excel-for-
business-and-financial-modelling-a-practical-guide-3rd-edition/
 Business Analytics Using
SAS Enterprise Guide and
® ®

SAS Enterprise Miner

® ™

A Beginner’s Guide

Olivia Parr-Rud

support.sas.com/bookstore
The correct bibliographic citation for this manual is as follows: Parr-Rud, Olivia. 2014. Business Analytics Using SAS®
Enterprise Guide® and SAS® Enterprise Miner®: A Beginner’s Guide. Cary, NC: SAS Institute Inc.
Business Analytics Using SAS® Enterprise Guide® and SAS® Enterprise Miner®: A Beginner’s Guide
Copyright © 2014, SAS Institute Inc., Cary, NC, USA
ISBN 978-1-62959-327-2
All rights reserved. Produced in the United States of America.
For a hard-copy book: No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in
any form or by any means, electronic, mechanical, photocopying, or otherwise, without the prior written permission of
the publisher, SAS Institute Inc.
For a web download or e-book: Your use of this publication shall be governed by the terms established by the vendor
at the time you acquire this publication.
The scanning, uploading, and distribution of this book via the Internet or any other means without the permission of the
publisher is illegal and punishable by law. Please purchase only authorized electronic editions and do not participate in
or encourage electronic piracy of copyrighted materials. Your support of others’ rights is appreciated.
U.S. Government License Rights; Restricted Rights: The Software and its documentation is commercial computer
software developed at private expense and is provided with RESTRICTED RIGHTS to the United States Government.
Use, duplication or disclosure of the Software by the United States Government is subject
to the license terms of this Agreement pursuant to, as applicable, FAR 12.212, DFAR 227.7202-1(a), DFAR 227.7202-
3(a) and DFAR 227.7202-4 and, to the extent required under U.S. federal law, the minimum restricted rights as set out
in FAR 52.227-19 (DEC 2007). If FAR 52.227-19 is applicable, this provision serves as notice under clause (c) thereof
and no other notice is required to be affixed to the Software or documentation. The Government's rights in Software
and documentation shall be only those set forth in this Agreement.
SAS Institute Inc., SAS Campus Drive, Cary, North Carolina 27513-2414.
October 2014
SAS provides a complete selection of books and electronic products to help customers use SAS® software
to its fullest potential. For more information about our offerings, visit support.sas.com/bookstore or call
1-800-727-3228.
SAS® and all other SAS Institute Inc. product or service names are registered trademarks or trademarks of SAS Institute
Inc. in the USA and other countries. ® indicates USA registration.
Other brand and product names are trademarks of their respective companies.
Contents

About This Book ...................................................................................... vii

About the Author ...................................................................................... xi
Chapter 1: Defining the Business Objective ............................................... 1
Introduction .................................................................................................................................... 1
Setting Goals .................................................................................................................................. 1
Descriptive Analyses ...................................................................................................................... 3
Customer Profile ...................................................................................................................... 3
Customer Loyalty ..................................................................................................................... 4
Market Penetration or Wallet Share ...................................................................................... 4
Predictive Analyses ........................................................................................................................ 4
Marketing Models .................................................................................................................... 5
Risk and Approval Models ...................................................................................................... 6
Predictive Modeling Opportunities by Industry .................................................................... 9
Notes from the Field .................................................................................................................... 13
Chapter 2: Data Types, Categories, and Sources ..................................... 15
Introduction .................................................................................................................................. 15
The Evolution of Data................................................................................................................... 16
Types of Data ................................................................................................................................ 17
Nominal Data .......................................................................................................................... 17
Ordinal Data ........................................................................................................................... 17
Continuous Data .................................................................................................................... 18
Categories of Data ....................................................................................................................... 18
Demographic or Firmographic Data .................................................................................... 18
Behavioral Data ...................................................................................................................... 19
Psychographic Data .............................................................................................................. 20
iv Contents

Data Category Comparison .................................................................................................. 20

Sources of Data ............................................................................................................................ 21
Internal Sources ..................................................................................................................... 21
Storage of Data ...................................................................................................................... 27
External Sources .................................................................................................................... 28
Notes from the Field .................................................................................................................... 29
Chapter 3: Overview of Descriptive and Predictive Analyses .................... 29
Introduction .................................................................................................................................. 29
Descriptive Analyses .................................................................................................................... 30
Frequency Distributions ........................................................................................................ 30
Cluster ..................................................................................................................................... 33
Decision Tree ......................................................................................................................... 33
Predictive Analyses ...................................................................................................................... 35
Linear Regression .................................................................................................................. 36
Logistic Regression ............................................................................................................... 39
Neural Networks .................................................................................................................... 41
Modeling Process ........................................................................................................................ 43
Define the Objective .............................................................................................................. 43
Develop the Model ................................................................................................................. 43
Implement the Model ............................................................................................................ 48
Maintain the Model ................................................................................................................ 49
Notes from the Field .................................................................................................................... 52
Chapter 4: Data Construction for Analysis ............................................... 53
Introduction .................................................................................................................................. 53
Data for Descriptive Analysis ...................................................................................................... 53
Data for Predictive Analysis ........................................................................................................ 54
Prospect Models .................................................................................................................... 55
Customer Models .................................................................................................................. 57
Risk Models ............................................................................................................................ 59
External Sources of Data ............................................................................................................. 61
Notes from the Field .................................................................................................................... 61
Chapter 5: Descriptive Analysis Using SAS Enterprise Guide ................... 63
Introduction .................................................................................................................................. 63
Project Overview .......................................................................................................................... 63
 Contents v

Project Initiation ........................................................................................................................... 64

Exploratory Analysis .................................................................................................................... 65
Importing the Data ................................................................................................................. 65
Viewing the Data .................................................................................................................... 66
Exploring the Data ................................................................................................................. 66
Segmentation and Profile Analysis............................................................................................. 69
Correlation Analysis ..................................................................................................................... 76
Notes from the Field .................................................................................................................... 77
Chapter 6: Market Analysis Using SAS Enterprise Guide .......................... 79
Introduction .................................................................................................................................. 79
Project Overview .......................................................................................................................... 79
Market Analysis ............................................................................................................................ 80
Project Initiation..................................................................................................................... 80
Data Preparation .................................................................................................................... 80
Penetration and Share of Wallet .......................................................................................... 89
Results .................................................................................................................................... 90
Notes from the Field .................................................................................................................... 91
Chapter 7: Cluster Analysis Using SAS Enterprise Miner .......................... 93
Introduction .................................................................................................................................. 93
Project Overview .......................................................................................................................... 93
Cluster Analysis ............................................................................................................................ 94
Initiate the Project ................................................................................................................. 94
Input the Data Source and Assign Variable Roles ............................................................. 97
Transform Variables .............................................................................................................. 99
Filter Data ............................................................................................................................. 102
Build Clusters ....................................................................................................................... 104
Build Segment Profiles ........................................................................................................ 107
Analyze Clusters and Recommend Marketing or Product Development Actions ........ 109
Notes from the Field .................................................................................................................. 109
Chapter 8: Tree Analysis Using SAS Enterprise Miner ............................ 111
Introduction ................................................................................................................................ 111
Project Overview ........................................................................................................................ 111
Decision Tree Analysis .............................................................................................................. 112
Initiate the Project ............................................................................................................... 112
vi Contents

Input the Data Source ......................................................................................................... 114

Create Target Variable ........................................................................................................ 115
Partition the Data ................................................................................................................. 117
Build the Decision Tree ....................................................................................................... 118
View the Decision Tree Output ........................................................................................... 120
Interpret the Findings .......................................................................................................... 126
Alternate Uses for Tree Analysis ........................................................................................ 128
Notes from the Field .................................................................................................................. 128
Chapter 9: Predictive Analysis Using SAS Enterprise Miner ................... 129
Introduction ................................................................................................................................ 129
Select ........................................................................................................................................... 130
Initiate the Project ............................................................................................................... 130
Select the Data ..................................................................................................................... 131
Explore ........................................................................................................................................ 133
StatExplore ........................................................................................................................... 133
MultiPlot ................................................................................................................................ 136
Modify .......................................................................................................................................... 138
Replace Missing Values via Imputation............................................................................. 138
Partition Data into Subsamples ......................................................................................... 139
Manage Outliers ................................................................................................................... 140
Transform the Variables ...................................................................................................... 142
Model ........................................................................................................................................... 145
Decision Tree ....................................................................................................................... 145
Neural Network .................................................................................................................... 147
Regression ........................................................................................................................... 148
Assess ......................................................................................................................................... 151
Notes from the Field .................................................................................................................. 155
References ............................................................................................ 157
About This Book

Purpose
This book serves as a tutorial for data analysts who are new to SAS Enterprise Guide and SAS
Enterprise Miner. It provides valuable hands-on experience using powerful statistical software to
complete the kinds of business analytics common to most industries. With clear, illustrated, step-
by-step instructions, it will lead you through examples based on business case studies. You will
formulate the business objective, manage the data, and perform analyses that you can use to
optimize marketing, risk, and customer relationship management, as well as business processes and
human resources.

Prerequisites
If you are a savvy business person with a desire to understand what drives your business, then this
book can help you get started. You need access to SAS Enterprise Guide or SAS Enterprise Miner
software; we provide you with example data to get started, but you will need data to analyze. An
understanding of basic statistics is helpful, but not required.

Organization
The book begins by helping you determine and structure the objective of your analysis in
accordance with the goals and objectives of your organization or department.

Chapter 2 describes types and sources of data for analysis. Chapter 3 offers an overview of
common business analyses, covering both descriptive and predictive analysis. Chapter 4 shows you
how to construct a data set for analysis. Chapter 5 details step-by-step instructions for a simple
descriptive analysis. Chapter 6 offers the same level of detail for a typical market analysis.
Chapters 7 and 8 offer a step-by-step guide to cluster and tree analyses, respectively. Each chapter
concludes with a section headed “Notes from the Field,” which offers related business advice and
leadership tips.

To conclude, Chapter 9 brings several concepts together in a full step-by-step case study for
building and comparing predictive models, culminating in final “Notes from the Field.”
viii

Examples
SAS Institute and SAS Press provide access to software updates and the author’s example data sets
so that you can practice the examples in this book.

Software Used
The software packages used in the writing of this book are SAS Enterprise Guide 6.1 M1 and SAS
Enterprise Miner 13.1. Although these are the latest versions available at the time of publication,
new features will appear in later releases. Visit the SAS Products and Solutions webpage for
updates and enhancements to all SAS system software at http://www.sas.com/en_us/software/all-
products.html.

Data Sets
You can access the data used in the author’s examples by linking to this book’s author page at
http://support.sas.com/publishing/authors. Select the name of the author. Then look for the cover
thumbnail of this book, and select Example Data to display the SAS data sets associated with this
book.

For an alphabetical listing of all books for which example code and data sets are available, see
http://support.sas.com/bookcode. To display a book’s example code, select its book title.

If you are unable to access data sets through the website, email [email protected].

Additional Help
Although this book illustrates many analyses regularly performed in businesses across industries,
questions specific to your aims and issues may arise. To fully support you, SAS Institute and SAS
Press offer you the following help resources:

• For questions about topics covered in this book, contact the author through SAS Press:

◦ Send questions by email to [email protected]; include the book title in your

correspondence.

◦ Submit feedback on the author’s page at http://support.sas.com/author_feedback.

• For questions about topics in or beyond the scope of this book, post queries to the relevant
SAS Support Communities at https://communities.sas.com/welcome.

• SAS Institute maintains a comprehensive website with up-to-date information. One page that
is particularly useful to both the novice and the seasoned SAS user is its Knowledge Base.
 ix

Search for relevant notes in the “Samples and SAS Notes” section of the Knowledge Base at
http://support.sas.com/resources.

• Registered SAS users or their organizations can access SAS Customer Support at
http://support.sas.com. Here you can pose specific questions to SAS Customer Support; under
Support, click Submit a Problem. You will need to provide an email address to which replies
can be sent, identify your organization, and provide a customer site number or license
information. This information can be found in your SAS logs.

Keep in Touch
We look forward to hearing from you. We welcome questions, comments, and concerns. If you want to
contact us about a specific book, please include the book title in your correspondence. For a complete
list of books available through SAS, visit http://support.sas.com/bookstore.

SAS Books
Reach our bookstore by phone, fax, or email:

• Phone: 1-800-727-3228

• Fax: 1-919-677-8166

• Email: [email protected]

SAS Book Report

Receive up-to-date information about all new SAS publications via email by subscribing to the SAS
Book Report monthly eNewsletter. Visit http://support.sas.com/sbr.
x
About the Author
Olivia Parr-Rud, an internationally recognized expert in predictive
analytics, business intelligence, and innovative leadership, founded the
SAS Data Mining Users Group, having been a SAS user since 1991 and a
SAS instructor and conference presenter for many years. She hosts the
popular VoiceAmerica Business radio show Quantum Business Insights
(http://www.voiceamerica.com/show/2240/quantum-business-insights) and
is a thought leader in the integration of analytic tools and leadership
practices to optimize performance and organizational agility. Her
predictive analytics research founded her first book, Data Mining
Cookbook: Modeling for Marketing, Risk and Customer Relationship
Management (Wiley 2000), which unveiled links between the global
economy and organizational dynamics, a topic featured in her second
book, Business Intelligence Success Factors: Tools for Aligning Your Business in a Global Economy
(Wiley and SAS Institute Inc. 2009). Her current research will develop a model that tracks the
alignment of business intelligence with human intelligence. Parr-Rud holds a B.A. in mathematics and
an M.S. in statistics. In addition to public speaking, she offers training and consulting in both
predictive analytics and business leadership, with clients that include numerous Fortune 500
companies. For more, visit http://oliviagroup.com.

Learn more about this author by visiting her author page at

http://support.sas.com/publishing/authors/parr-rud.html. There you can download free book excerpts,
access example code and data, read the latest reviews, get updates, and more.
xii
Chapter 1: Defining the Business Objective
Introduction ........................................................................................................ 1
Setting Goals ....................................................................................................... 1
Descriptive Analyses ........................................................................................... 3
Customer Profile ................................................................................................................. 3
Customer Loyalty ................................................................................................................ 4
Market Penetration or Wallet Share ................................................................................. 4
Predictive Analyses ............................................................................................. 4
Marketing Models ............................................................................................................... 5
Risk and Approval Models ................................................................................................. 6
Predictive Modeling Opportunities by Industry ............................................................... 9
Notes from the Field ......................................................................................... 13

Introduction
In today’s highly competitive global economy, organizations are increasingly dependent on their
ability to leverage accurate, accessible, actionable data. This increased dependence on data requires
a new kind of visionary leadership focused on a smart investment in an enterprise data management
system, the right analytic talent to leverage it, and a culture designed to support “big data.”

A culture of big data includes powerful analytic tools. Advanced analytics and reporting software
platforms, such as SAS Enterprise Guide and SAS Enterprise Miner, enable the user to extract deep
insights from big data and its underlying patterns. This book is designed to guide the business
analyst or manager as he or she seeks these insights by providing case studies and everyday
examples.

In this chapter, you will learn the importance of clearly defining the objective of any data analysis
project.

Setting Goals
The use of data analysis is now commonplace in business across industries. Many applications,
such as customer profiles or response or risk models, are quite straightforward. However, as
companies attempt to develop models that address more complicated measures such as customer
2 Business Analytics Using SAS Enterprise Guide and SAS Enterprise Miner

retention and lifetime value, the importance of clear goals is magnified. Failure to define the goal
correctly can result in wasted dollars and a lost opportunity.

The first and most important step in any data analysis project is to establish a clear goal, not a goal
defined only by the data or the method, but a goal that makes sense to the business as a whole. In other
words, the goal of the analysis must be defined in terms of how it will help the business reach its
strategic goals. When you engage the stakeholders, questions and analytic methods will likely include
the following:

• Do you need to understand the characteristics of your current customers? This goal may
involve looking for averages or segmenting your customer base and creating profiles.
• Do you need to attract new customers? Response modeling on customer acquisition
campaigns enables you to lower the marketing costs of attracting additional customers.
• Do you need to avoid high-risk customers? Risk or approval models identify customers or
prospects that have a high likelihood of incurring a loss for the company. In financial services,
for example, a typical loss comes from nonpayment on a loan. Insurance losses result when
claims filed by the insured outweigh the calculated loss reserves.
• Do you need to make your unprofitable customers more profitable? Cross-sell and up-sell
target models can be used to increase profits from current customers.
• Do you need to retain your profitable customers? Retention, or “churn,” models identify
customers with a high likelihood of lowering or ceasing their current levels of activity. The
identification of these customers before they change their behavior enables the construction of
strategies and actions to retain them. The cost of retaining a customer is often less than the
cost of winning back a customer.
• Do you need new customers to be profitable? A lifetime value model identifies prospects with
a high likelihood of being long-term profitable customers. Combining several measures, such
as response, activation, and attrition, can help you target prospects that have the highest value
over the life of the product or service.
• Do you need to win back your lost customers? Win-back models are built to target former
customers.
• Do you need to improve customer satisfaction? In today’s competitive market, customer
satisfaction is crucial to success. Combining market research with customer profiling is an
effective method of measuring customer satisfaction.
• Do you need to increase sales? You can increase sales in several ways. A new-customer-
acquisition model can grow the customer base, leading to increased sales. Cross-sell and up-
sell models can also be used to increase sales.
• Do you need to determine what products or services to bundle or offer sequentially? Affinity
analysis can identify products that have a high probability of being purchased together or
within a narrow time frame.
• Do you need to reduce expenses? Better targeting through the use of models for customer
acquisition and customer relationship management can improve the efficiency of your
marketing efforts by reducing expenses.
 Chapter 1: Defining the Business Objective 3

• Do you need to determine the most effective channel or sequence of channels? Analysis of
purchase behavior can identify and measure which channel or sequence of channels is most
efficient.
• Do you need to optimize the frequency of your offers? Analysis based on frequency testing
can indicate how often to send an offer. This frequency strategy may vary greatly by channel
in terms of both cost and results.
• Do you need to deliver the right message to the right person at the right time through the
right channel? This objective is every marketer’s overall goal and encompasses all of the
more specific goals.

So far, the list of goals concerns selling a product or service. Other business uses of data analysis
that can result in cost reductions and higher profits may involve answering the following related
questions:

• Do you need to avoid process failure? Companies in a variety of industries are modeling data
collected from production lines, customer feedback, hospital error, and similar sources to
predict breakdowns and take corrective actions.
• Do you need to analyze health treatments? Predictive models in the life sciences have saved
money and lives. Models that predict disease outbreaks are used to calculate inventory in
high-risk geographic regions.
• Do you need to optimize your inventory? Knowing how many of what products to have on
hand in a specific location can save time and money.
• Do you need to optimize your staffing? Demand can be predicted, which enables stores,
resorts, and other businesses to know exactly whom to have on-site or available for work.
• Do you need to predict the best location for a future store, restaurant, or other business?
Planning growth can be facilitated by the prediction of sales for a specific geographic area.
For numerous other examples of goals, see Predictive Analytics (Siegel and Davenport 2013).

To achieve your goals, you will choose appropriately from a variety of analyses. The two major
categories of analysis are Descriptive Analysis and Predictive Analysis.

Descriptive Analyses
Descriptive analysis is a technique that allows you to view and measure your company and
customer characteristics. The results of the analysis can be used to guide decisions in every area of
your company. Therefore, descriptive analysis is an essential first step to managing your business.

Customer Profile
Because your best customers typically drive your company’s profitability, analyzing your customer
base is a good first step. This provides a snapshot of exactly who is buying your products or
services.
4 Business Analytics Using SAS Enterprise Guide and SAS Enterprise Miner

A customer profile analysis identifies and measures the characteristics of your most profitable
customers. Insights gained from this analysis can be used to enhance and customize your marketing
activities. Customer profile analysis involves measuring common characteristics within a
population of interest. Demographics, such as average age, gender (percent male), marital status
(percent married, percent single, and so on), and average length of residence, are typically included
in a profile analysis. Other measures might be more business-specific, such as the age of a
customer relationship or the average risk level.

Also, if you find distinct characteristics that define a particularly profitable segment, you can match
those characteristics to names on other databases or external lists and purchase new names for
marketing. Similarly, if you identify the most and least profitable customers, you can focus your
retention efforts and offer varying levels of customer service.

Customer Loyalty
Loyalty is an aspect of the customer profile that is measured by the length of the relationship, the
amount of spending, or a combination thereof. Unlike customer profile analysis, which is more of a
snapshot of your base, customer loyalty analysis looks at your customer base over time. History
tells us that, on average, loyal customers bring in more revenue at a lower cost.

Market Penetration or Wallet Share

Market penetration analysis and wallet share analysis are techniques for measuring the
performance of your customer base in comparison with the performance of the overall market for
your industry. These analyses typically look at these measures across selected demographics or
other characteristics, or a combination thereof.

Your company’s market penetration compares the number of customers in your base to the possible
customers in the overall market. Wallet share compares your revenues to those available in the
market. It is powerful for determining how much potential lies with a specific segment as defined
by a demographic or behavioral characteristic.

Wallet share analysis can be used to prioritize marketing dollars in industries where the customer
has a fixed amount to spend with multiple businesses (their “wallet”). You calculate it by taking the
amount that the customer is spending with your company and dividing it by the total that they have
to spend with you and your competitors.

Predictive Analyses
Predictive analysis uses statistics, machine learning, or data mining to develop models that predict
future events on the basis of current or past data. In business, there are numerous models that are
commonly used.
 Chapter 1: Defining the Business Objective 5

Marketing Models
Marketing models are models that are built and used specifically for growing your business.

Response
A response model is typically the first type of target model that a company seeks to develop. If no
targeting was done in the past, a response model can provide a significant boost to the efficiency of
a marketing campaign by increasing responses or reducing expenses. The objective is to predict
who will respond to an offer for a product or service. It can be based on past behavior of a similar
population or some logical substitute. For more detail on data for modeling, see Chapter 2.

Win-Back
A win-back model is used to invite former customers to reconsider their relationship to the
business. These models can be powerful, because you already have good behavioral data on these
former customers. You should consider risk and retention measures when appropriate.

Activation
An activation model predicts whether a prospect will become a customer. It is applicable only in
certain industries, most commonly credit card and insurance. For example, for a credit card
prospect to become an active customer, the prospect must respond, have their credit approved, and
use the account. If the customer never uses the account, he or she actually costs the bank more than
a noncustomer. Most credit card banks offer incentives, such as low-rate purchases or balance
transfers, to motivate new customers to activate. An insurance prospect can be viewed in much the
same way. A prospect can respond and be approved, but if he or she does not pay the initial
premium, the policy is never activated.

Revenue
A revenue model predicts the dollar amount of an expected sale. This model is useful for
distinguishing low-value from high-value responders or customers. For example, if a prospect
becomes a customer after buying a product online, he or she has a specific value based on the
amount of that purchase.

Usage
A usage model predicts the amount of use given to a product or service. This model is most
applicable to the telecommunications companies that determine their profits by estimating minutes
of usage. Life sciences companies predict disease outbreaks in specific geographic locations, on the
basis of doctor visits, sales of prescriptions, and over-the-counter drugs.

Cross-Sell and Up-Sell

A cross-sell model is used to predict the probability or value of a current customer’s buying a
different product or service from the same company. An up-sell model predicts the probability or
value of a customer’s buying more of the same product or service.
6 Business Analytics Using SAS Enterprise Guide and SAS Enterprise Miner

Selling additional products or services to current customers is quickly replacing new customer
acquisition as one of the easiest ways to increase profits. Testing offer sequences can help to
determine what the next offer should be and when to make it. This testing enables companies to
carefully manage offers to avoid over-soliciting and possibly alienating their customers.

Loyalty, Attrition, and Churn

A familiar mantra in marketing is, “It costs less to retain a customer than it does to replace one.”
Every company benefits from loyal customers. Loyalty, also known as retention, can be thought of
as the opposite of attrition—the loss of a customer relationship. For highly competitive industries
where markets are close to saturation, these terms are particularly relevant. Industries that count on
renewals, such as insurance, telecommunications, and publishing, pay close attention to retention
and attrition.

Attrition is defined as a decrease in the use of a product or service. For credit card banks, attrition is
the decrease in balances on which interest is earned. It occurs when customers switch companies,
usually to take advantage of a better deal. For years, credit card banks lured customers from their
competitors by offering low interest rates. Telecommunications companies continue to use strategic
marketing tactics to lure customers away from their competitors. Many other industries spend much
effort to retain customers and steal new ones from their competitors.

For several decades, credit cards have been offered to almost every segment of the population,
resulting in a saturated market. This reality means that credit card banks are now forced to increase
their customer base primarily by luring customers from other providers. Their tactic is to offer low
introductory interest rates for anywhere from three months to one year or more on either new
purchases or balances transferred from another provider, or both. Their hope is that customers will
keep their balances with the bank after the interest converts to the normal rate. Many customers,
though, are savvy about keeping their interest rates low by moving balances from one card to
another before the rate returns to normal. These activities introduce several modeling opportunities.
One kind of model predicts the customer’s reduction or ending of the use of a product or service
after an account is activated.

In some industries, attrition becomes cyclical and is known as churn, which is defined as
customers’ closing of one account in conjunction with the opening of another account for the same
product or service, usually at a reduced cost. Companies have learned that it is cheaper to keep a
customer happy than to replace him or her. To this end, retention and attrition modeling have
become popular in some industries.

Businesses use social network analysis to predict churn by examining the group behavior. By
measuring the density of connection in a mobile phone network, for example, cellular companies
are able to predict patterns of churn and take proactive steps to retain their customers.

Risk and Approval Models

Risk models, or approval models, are unique to industries that assume the potential for loss when
offering a product or service. The most familiar kinds of risk occur in the banking and insurance
 Chapter 1: Defining the Business Objective 7

industries. However, there is a new focus on models to predict risk in the form of fraud and security
breaches. As a result of the growth of data and interconnectivity, the level of risk related to data
and technology is increasing exponentially.

Default
Default models have been used to determine risk for several decades. On the basis of an
individual’s or business’s credit history, financial profile, demographics, or similar data, a lending
institution or other business can determine its applicant’s likelihood of meeting his or her loan
obligation. Predicting the likelihood of default or bankruptcy has been a main profit-driver of the
loan industry for many years. Whether it’s for a mortgage, credit card, car loan, or even utility
service, many companies rely on a credit score to determine approval and terms of a contract.
Banks also use aggregations of the expected loss to meet regulatory requirements for loan reserves.

Loss-Given-Default
At the point of default, loss-given-default models are used to estimate the size of the loss. Some
loans, such as mortgages and automobile loans, are secured, meaning that the bank holds the title to
the home or automobile for collateral. The risk is then limited to the loan amount minus resale
value of the home or car. Unsecured loans are loans for which the lender holds no secured assets,
such as credit cards.

Many other industries incur risk by offering a product or service with the promise of future
payment. This category includes telecommunications companies, energy providers, retailers, and
many others. The type of risk is similar to that of the banking industry in that it estimates the
probability of a customer’s defaulting on the payment for goods or services.

Claim
For the insurance industry, the risk is that of a customer’s filing a claim. The basic concept of
insurance is to pool risk. Insurance companies have decades of experience in managing risk.
Insurance companies use predictive model scores to determine approval, optimize pricing, and
determine reserve levels for life, health, auto, and homeowners insurance on the basis of
demographics, risk factors including claim history, and credit risk factors. Because of heavy
government regulation of pricing in the insurance industry, managing risk is critical to insurance
companies’ profitability. Warranty companies often use product information and safety records to
model the likelihood of a claim.

Fraud
With the increase in electronic and online transactions, fraud is increasing at an alarming rate.
According to fraud research by the 2012 Global Fraud Study, survey participants estimated that the
organizations lose an average of 5% of their revenues to fraud each year (Association of Certified
Fraud Examiners 2012). Applied to the 2011 gross world product, this figure translates to a
potential projected annual fraud loss of more than $3.5 trillion. Because of fraudulent purchases
made with stolen credit cards, fake insurance claims (including life, health, and auto), stolen
8 Business Analytics Using SAS Enterprise Guide and SAS Enterprise Miner

cellular usage, false tax returns, account fraud through Automated Clearing House, and money
laundering, fraud increases costs for businesses and consumers alike.

Powerful models are in use and in development to thwart these costly actions.

Credit card banks use predictive models to identify the types of purchases that are typically made
with a stolen card. For example, if the card is used to purchase expensive jewelry, furs, or firearms,
the bank will freeze the card until the cardholder can verify the charges.

Insurance claim fraud is also costly to insurers and policyholders. Historically, every claim was
personally inspected for fraud. Today, predictive models can estimate the likelihood that a claim is
fraudulent based on common characteristics of past fraudulent claims. This model allows insurers
to reduce expenses by automating the approval of low-risk claims and placing a priority on claims
with a high likelihood of fraud.

Government institutions, including the Internal Revenue Service and social service agencies such
as Medicare and the Social Security Administration, use predictive models to identify false tax
returns or requests for benefits.

Lifetime Value
Lifetime value (LTV) is the expected value of a prospect or customer over a specified period of
time, measured in today’s dollars. When calculated with the use of a series of models or other
estimates, it allows a company to make decisions that optimize overall profits.

LTV is measured in various ways, depending on the industry, but basically it represents the value
of future revenues minus overhead and expenses. It can be estimated from a series of estimates
from models or other sources that may consider one or more of the following: response, activation,
retention, risk, cross-sell, up-sell, and revenue. This evaluation enables companies to allocate
resources based on customer value or potential customer value.

Historically, marketing strategies were driven by the financial benefits of a single campaign.
Customer profitability was optimized by the net profits of the initial sale. With the increased cost of
acquiring customers and the expansion of products and services to existing customers, companies
are expanding their marketing strategies to consider the LTV of a potential customer.

LTV measurements on a customer portfolio can quantify the long-term financial health of a
company or business. However, focusing on long-term goals can be challenging for companies if
they suboptimize short-term profits.

The customer life cycle comprises three main phases (Figure 1.1):

• prospect
• new or established customer
• former or lapsed customer
 Chapter 1: Defining the Business Objective 9

Many opportunities for developing predictive models exist within the life cycle.

Figure 1.1: Customer Life Cycle

A prospect can be modeled for the propensity to respond or to activate. The risk level can also be
estimated with predictive modeling or segmentation. By combining response, activation, or risk
models with some customer value estimates such as retention (churn) or subsequent revenues,
companies can calculate a prospect’s LTV.

After a prospect becomes a customer, numerous additional modeling opportunities exist. Finally,
after the customer relationship ends, models can be developed to improve profitability. By
integrating models into a customer relationship management program, companies can reduce
expenses while improving the efficiency of their marketing efforts.

Predictive Modeling Opportunities by Industry

In the past few decades, predictive models have gained popularity across a variety of industries.
Statisticians and data miners have found that the methods pioneered by the financial services
industry have broad applicability to most other industries.

For all of the industries listed in this section, a combination of factors drives profitability. The
financial impact of these factors can be predicted with the following types of models:

Banks: Credit Card, Retail, Mortgage, and Savings and Loan

• response model—to target offers for loans
• default model—to predict non-payment or bankruptcy
10 Business Analytics Using SAS Enterprise Guide and SAS Enterprise Miner

• revenue model—to estimate the amount of purchases and earned interest

• retention/loyalty/attrition model—to estimate the optimal combination of pricing and services
• cross-sell/up-sell model—to increase usage on credit cards or purchase additional products
and services such as investment products, retail banking services (savings and checking
accounts), refinance of mortgage, etc.
• fraud model—to detect lost or stolen cards or identity theft
• collection model—to prioritize resources in the collection of debt

Insurance Companies: Life, Health, Auto, Property, and Casualty

• response model—to target offers for insurance
• default model—to predict likelihood of claims
• retention/loyalty model—to estimate the optimal combination of pricing and services
• cross-sell/up-sell model—to increase insurance coverage (life) or purchase other types of
insurance
• fraud model—to predict the likelihood of a false insurance application or claim

Retailers: Catalog Companies and Retail

• response model—to target offers for goods and services
• catalog companies and offline retailers—to reduce snail mail costs
• online retailers—to improve targeting
• revenue model—to estimate the amount of the purchases
• retention/loyalty model—to estimate the optimal combination of marketing actions and
incentives
• cross-sell/up-sell model—to increase repeat sales

Telecommunications Companies: Cable, Internet, Wired-phone, and Cellular

• response model—to target offers for phone or internet service
• revenue model—to estimate the amount of usage
• retention/loyalty model—to estimate the likelihood of cancellation or churn
• default model—to predict the likelihood of defaulting on payment
• cross-sell/up-sell model—to increase usage and optimize product and service bundling
• fraud model—to predict illegal access to services
• process failure model—to predict grid or supply-chain failure

Entertainment and Social Media Industry: Radio, Television, Social Media, and
Online Sharing Sites
• response model—to increase click-through, views, and purchases based on variations in ad
content, timing, and placement
 Chapter 1: Defining the Business Objective 11

• revenue model—to estimate the amount of the purchases

• retention/loyalty model—to estimate the optimal combination of marketing actions and
incentives
• failure model—to predict internet or media delivery interruptions

Gaming Industry: Casinos and Online Gaming

• response model—to optimize ad placement and incentives
• retention/loyalty model—to estimate the optimal combination of marketing actions and
incentives; for example, many casinos have strong loyalty programs that include food,
lodging, and priority access
• cross-sell/up-sell model—to increase usage and optimize product and service bundling
• fraud model—to predict the likelihood of a false representation, such as a stolen credit card or
identify theft

Education: Colleges, Universities, Technical Colleges, and Art Schools

• response model—to target an offer based on the likelihood of acceptance
• retention model—estimate the likelihood of graduation

Technology Companies: Computer Hardware and Software

• response model—to target offers for product or service
• revenue model—to estimate the amount of usage for services
• retention/loyalty model—to estimate the likelihood of repeat purchases
• cross-sell/up-sell model—to increase usage and optimize product and service bundling
• process failure model—to predict product, operations, or supply-chain failure

Utilities: Gas, Electric, and Water

• response model—to target offers for services in deregulated areas
• usage model—to estimate the amount of usage.
• retention/loyalty model—to estimate the likelihood of cancellation or churn
• default model—to predict the likelihood of defaulting on payment
• cross-sell/up-sell model—to increase usage and optimize product and service bundling
12 Business Analytics Using SAS Enterprise Guide and SAS Enterprise Miner

• fraud model—to predict illegal access to services

• failure model—to predict grid or supply-chain failure

Hospitality and Travel Industry: Hotels, Resorts, and Restaurants

• response model—to target offers for phone or internet service
• usage model—to estimate the number of potential customers in order to plan staffing and
supplies
• revenue model—to estimate the amount of revenue
• retention/loyalty model—to estimate the likelihood of repeat customers
• cross-sell/up-sell model—to increase usage and optimize product and service bundling
• process failure model—to predict operations or supply-chain failure

Manufacturing: Auto, Pharmaceutical, Biotech, Chemical, Food, etc.

• response model—to predict sales
• usage model—to predict product expiration, disease outbreak, poisoning, allergic reaction
• revenue model—to predict production needs
• process failure model—to predict a breakdown in operations or a supply-chain
• product failure model—to predict the likelihood of a product malfunction or safety recall

Public Sector and Nonprofit: Government, Political, Religious, Community,

Education, Healthcare
• response model—to predict who will donate or help with campaign
• revenue model—to estimate the amount of donations
• target model—to predict for whom someone will vote
• fraud models—to detect false charges for services, identity theft, or IRS tax filings
• usage models—to plan staffing and supplies for hospitals, emergency centers, first
responders, or nursing homes

Transportation and Shipping: Rail, Ship, Truck, and Airline

• usage model—to predict quantities, sales, and fuel costs
• geo usage model—to optimize routes and storage locations
• process failure model—to predict operations or supply-chain failure

Sports
• recruitment model—to predict whether a player’s statistics are a good fit for the team
• target model—to predict who will attend an event
Discovering Diverse Content Through
Random Scribd Documents
had theirs and we both of us did it. The children of the neighbors
was often round with us then, but they made us think of 'ourn that
was to come, in the favor of the Lord. And if when I helped her
along from plank to plank, I held her hand a little longer than
absolutely necessary, who was to care.
"There's been no decided change in the years; we've been honey-
mooning along just about the same. Course with the children she
had more to do in doors, but she's always managed, if there was an
extra run o' fish to come to the flakes and help me over the rush;
and if one o' the kids was sick or anything extra come, why I've
always toted the load for her."
During the last few sentences Jim was watching the clock intently
and as he spoke the last sentence, he crossed the little room and
began winding the clock. I looked up and there, sticking out from
behind the clock, was a worn, brown wallet. Evidently he was still
living up to his habit of sharing.
"It's time all decent folks was in bed," he said. "We done want to
ape the city folks."
So bidding them good night I went out into the night. The rain had
ceased and there were fast hurrying clouds breaking up and I could
see the moon high over the spruces. I felt my way along the road
back to St. John's.
 CHAPTER X
JIM AND HIS BOOK

"T
hey that seek the Lord understand all things." Jim spoke
with his usual deliberation. Again, I had found my way to
the little house, where now I felt welcome. It was "lightin'
an' readin'" time as Jim called it.
"They that seek the Lord, understand all things," repeated Jim. I'm
finding it true more and more. It is true that the Lord giveth to a
man what is good in His sight, wisdom and knowledge and joy.
"We began sharin' the book, just as we began sharin' the wallet. I
callated that since the Lord by wisdom founded the earth we'd have
to found our earth home the same way.
"I'm not educated with figgering knowledge. I never got much
school wisdom, for I never went much, and what I did get was
mostly from the fellow that set on the bench with me instead of
from the teacher. The teacher was so busy with fifty odd pupils,
varying from four to twenty years in age, that he didn't have much
time for any one. He had to skip from the multiplication table to
algebra and often he skipped some of the pupils, and I was apt to
be the one he overlooked.
"I know my limitations. A city chap told me about them once and I
thanked him." Jim chuckled at the remembrance.
"'Look ahere,' the city chap said to me, 'do you know you've lost all
the G's out of your vocabulary. Your words don't look nor sound
natural. You better start in putting them on. And there is no such
word as ain't. Remember that or you can't talk in polite society.'
"I presume he knew, for he talked as though he was on good terms
with a dictionary; and when he went fishing and caught the hook in
his hand he said words that weren't in the dictionary, and that came
near breaking the first commandment. I've got some of those G's
put back on, but not all. Two things is helping me on the job, the
reading of the Good Book and the children.
"Book learnin's a fine thing. I'm stumblin' along thru a book or two
myself, but I callate the prophets didn't refer to book knowledge
when they wrote of wisdom, but rather heart and soul wisdom. The
promise I recollect was this: 'For wisdom shall enter into thine heart
and knowledge shall be pleasant unto thy soul.'
Then he reached for his Bible, but before he opened it he said:
"This is the most valuable thing in this house. I've been in houses in
St. John's fussed up with furniture and things, so many you felt you
would disturb 'em by setting down, but this book wasn't no where to
be seen and once I asked a woman to let me look at the Book, and
she said she'd have to keep me waitin' till she found it, but she was
quite sure she had it. Guess its wisdom never got very far into her
soul.
"It's a satisfyin' book. Readin' of it is like quenching your thirst at a
hill spring. In the days afore I was converted as a young fellow with
the rest, I used to sail over to the French Island of St. Pierre and
smuggle in a few gallons of rum. But it never quenched my thirst. It
would leave me afterward, all-fired thirsty. But open this book and
you find fountains of cool water.
"I've tried in the years to halt at the springs as Moses and his people
did when they crossed the desert and come to a spring. There's
many a river of the water o' life flowing sweet and fair as we journey
thru the good book, but to me the promises are the springs and wife
and I have lingered longest at the springs. We've marked them and
there's a good many of them and we haven't found them all yet. She
has helped me mark 'em. A fisherman's hands get a bit calloused
and clumpsy and she does most of the markin', but I do my share of
the discoverin'. It's always a happy night, when we find a new spring
and rejoice in a new promise, but it's a glad night when we quench
our thirst at any one of the never-failing springs. Their all of 'em
fresh an' sparklin'; there's nary a one of His that are salt or bitter.
"Effie keeps a pencil handy there with her sewing things and when I
find a new promise, I hand over the book to her and she underlines
it. Then the favorite springs we mark in the margin, so we'll find 'em
easy as we journey."
He opened the book, his book it was in more ways than one. It was
very much worn; its leaves were thumbed and now and then as he
turned the pages a fish scale dropped out.
"Here are the Great Mountain Springs. The Master indicated them
with a big, Blessed, so we wouldn't miss them, perhaps the clearest
one is this, Blessed are the pure in heart, for they shall see God, but
they've all got sparkling water; their all promises that quench the
soul's thirst.
"You will find some of these same markers in the Old Testament,
though few folks seem to search there for the Blesseds. Here are
some of the springs that are marked for our use.
"'Blessed are they that wait for Him!'
"'Blessed is he whose transgression is forgiven, whose sin is
covered.'
"'Blessed is the man whom thou choosest and causest to approach
unto Thee, that he may dwell in Thy courts: we shall be satisfied
with the goodness of Thy house, even of Thy holy temple.'
"'Blessed is he that considereth the poor.'
"'Blessed is the man whose strength is in Thee, in whose heart are
the highways to Zion.'
"Let me turn the pages slowly and when I come to a favorite spring
we'll halt a moment," commented Jim as he continues his reading.
"'He shall give his angels charge over thee, to keep thee in all thy
ways.'
"It won't hurt you a mite, if you hev to wait awhile atween the
verses. Most parsons read the Bible too fast. They go scurryin' thru
the readin' like as though a shower was comin' an' they had to get in
out of it post haste."
"'Fear not; I am with thee; be not discouraged; for I am thy God; I
will strengthen thee, yes, I will help thee; yea, I will uphold thee
with the right hand of my righteousness.'
"'With long life will I satisfy him and show him my salvation.'
"That there first part has puzzled me somewhat, for I've known
many a one to die young. My folks used to say the good died young,
cause the Lord had need of 'em over there. Struck me as kinder
queer. But I reckon He meant here just what He said, as He does
elsewhere. It's His intention to have long life and goodness go
together, only some of us interferes with His plan, but He lets us
interfere 'cause it's best and will work out His way in the end."
"'He shall call upon me and I will answer him. I will be with him in
trouble. I will deliver him and honor him.'
"'Behold I will bring thee health and a cure.'
"'The Lord shall be thine everlasting light, and the days of thy
mourning shall be ended.'
"'There shall no evil befall thee, neither shall any plague come nigh
thy dwelling.'
"Did you think," said Jim, interrupting his reading, "that there were
so many bright, clear springs for the traveller?"
Then, without waiting for any answer, he continued slowly turning
the pages, reading me from his marked places.
"'Delight thyself in the Lord and He shall give thee the desires of thy
heart.'
"'The joy of the Lord shall be your strength.'
"'He that endureth to the end shall be saved.'
"There are signs put up, too, not only to mark springs but to inform
us," interpolated Jim.
"Now once as we was journeying, it come over me that these
springs may have been intended for others and not for us and that
very night, I come upon this sign and it took every bit of doubt out
of my heart.
"'For the promise is unto you and your children.' How could it be
plainer than that?"
As he closed the Book I said: "I, too, have a Book but I think
sometimes I have lost my way as I journeyed and I am going to put
up sign-boards of my own now, so I'll never lose my way again.
There is no use to camp in dark valleys when just beyond are the
hills and the springs. It's unwise to wander thru deserts of
generalizations when the promises are close at hand."
"Yes," added Jim, "what do we care whether King Agag was hewed
to pieces or not. We know the words of salvation."
 CHAPTER XI
RAILROADING WITH THE KID

I
f there is anything that I have told you about Harbor Jim that
sounds feeble or sickly sentimental, I have told you an untruth.
Turn back to where I said it, and cross it out. It doesn't belong in
this story. It's rank injustice to Jim.
I have fished with a good many of the Landers. I have been fishing
off the banks when the weather has kept every man of us praying,
who knew how to pray, and I have had a chance to judge of these
bronzed fellows, big of hand and foot and the same of heart, most
of them, as they met the wind and weather, the fortunes of life on
the sea and the shore; and I want you to know I never have known
and loved a manlier man than Jim.
Maybe that was why I was surprised one morning as we were
returning to camp from a trip up the Humber River after salmon, to
see the tears rolling down his cheeks and to note that he hastily
took his sleeve and wiped his face and swallowed hard.
In this land of uncrossed lakes and unfished rivers, there is probably
not a fairer one than the Humber River and there are reminders of
Norway both on the lower and upper Humber.
It was with some difficulty that I had persuaded Harbor Jim to leave
his home for the trip inland to the Humber for salmon fishing. The
Lander does not take readily to a vacation, indeed, the average
Lander cannot afford to take one. After several days of argument,
Jim gave in, with this sentiment:
"I think the Lord must a been a good fisherman, else He wouldn't a
picked fishermen to follow Him. He wanted to swap stories with 'em
now and again. The Master knew by the ruffle and the shadow on
the lake when the fish was schooling and he told Peter where to let
down his nets. I have an idea He went away sometimes to fish as
well as to pray and that fishin' with Peter and John, they come to
know each other better."
After that Jim was as keen as a boy to get ready the lines and the
flies and to pack our little outfit. We went on the train to Deer Lake,
crossed the lower end of the lake and went up the river. We fished
near Steady Brook Falls and away up at Big Falls and the weather
was all that could be desired. We caught more salmon and trout
than we needed and we were bringing out all that the law would
allow us to transport. It had been the best week's fishing I had ever
had, and there had been some surprises. We had by chance fallen in
with an old friend of mine from the States and another day we had
seen a stag of great size following the birds down to a pool.
All had gone so well with us that I was at a loss to account for this
sudden demonstration of feeling. It was not like Jim. I knew him and
his way well enough, to know that he would not wish to be
questioned, so we tramped on in silence over the carry, and it was
not for an hour afterward that he ventured an explanation.
"There at the carry you may have thought it strange, the way I
acted up. That little fellow we seed there playing with his father's
canoe, made me think of little Peter. I've never mentioned him, I
seldom do, but I think a good deal about him and often I believe he
is with me. He made the carry and past over to Kingdom Come three
years ago.
"Do you know sometimes when I used to watch my little Peter
playing, and the light and shadows would be around him, I used to
think, pardon me, he looked like the pictures I've seen of the
carpenter's Son, His Son. He was our first child, born out of our first
wonderful love, but he never was a strong child. I don't know why. I
never could think of him as becomin' a fisherman. He used to like
better'n the average child, to journey with us thru the land o' the
springs during the evenings, and I thought mebbe the Lord would
call him to be a preacher, though I never let on to him, what I was a
thinkin'.
"When he was eight, he got kinder spindlin' and at the same time he
wanted to go to the woods and to see the island. He had another
hankering, too, that was to ride on the trains. He used to collect
engine numbers any time he was in St. John's. His mother used to
say that she believed he'd be an engine driver instead of a preacher.
"At first I didn't pay much attention when he asked to go, but as he
got thinner and paler, I began to take trips with him on the railroad.
We had great times together going to places and for a time they
seemed to chuck him up a bit. We went down to old Placentia one
time. Ever down there? It's a lovely old place; lies sprawled out on a
sandy beach with arms reaching round it and the hills sending down
beauty on to it. We climbed the hill across the gut from the town,
Castle Hill, and saw the crumbling ruins of the old French Fort and
we went across to Bradshaw's and saw the Communion set that was
presented by King William the Fourth.
"Sometimes we would take mother along and go to Top-sails and
look down the bay as we ate our lunch. Then one time we went over
to Belle Isle and saw the men working in the iron mines under the
sea and Peter talked about what he saw for weeks. I was worried a
good deal about him, but we both felt better on the trips. There was
always something to see. For miles our railroad gives you Conception
Bay with now a frame of hills and now one of spruces. Then in the
centre of the island are great lonesome barrens where the caribou
come to feed and the little nameless lakes are clustered. Peter had
'em all named, but I think he used to change the names sometimes.
There were so many his little mind forgot the long list.
"Then 'twas fun to be on our railroad. It's a road that throws you
about some; makes an impression on you, and a good hard one,
sometimes. But it's the only railroad we've got in the Dominion and
without it our country wouldn't have the farms it has now, nor
friends like you, coming and going.
"I remember when we took the sleeper, the kid and I. We didn't
often do that; we couldn't afford it, but this time we were going over
to the Codroys and I put the little fellow to bed and sat down for a
spell of thinking, across the aisle from him. Suddenly the train gave
a lurch. Guess the engineer got kinder hot stoppin' to drive cows off
the track and we was a hittin' it up as much as thirty miles an hour.
What do you think? Little Peter come a flyin' down from his berth
right into my arms and he says, not hurt a bit, only tickled:
"'Pa, a fellow has to be put to bed more'n once to stay put on our
road.'
"He always called it our road, though he knew its short-comings as
well as I.
"We only took one winter trip and that was a long one and I blamed
myself many a time for taking the risk, though I don't know's it hurt
him any, and I'm sure I always kept him warm and covered. When
we got to Gaff Topsails, the track ahead was solid, sheer ice and the
wind swept fierce across it from the south. They strapped the train
on the track, so's it wouldn't tumble over. Seems funny now, but it
wasn't then. But we didn't suffer any. They had lots of food-stuff
aboard and when it give out the train hands went across the snow to
the next town to get more. It took us fifteen days to get to Petrie's.
The store-keeper at Petrie's had been up to St. John's to buy goods
and he was on the train with us, anxious to get home. He was kind
to little Peter and rode him pig-a-back every day, when it was too
bad for him to walk about.
"The store-keeper reached Petrie's in thirteen days, two days ahead
of the train, by walking the last ten miles. His folks was surprised,
for they didn't expect him until the train got in.
"Still that trip we made better time than the trains sometimes do in
the winter. One train took twenty-six days to get across the island.
"On these trips, Peter and I would come home with many a story to
tell mother and little Peter would be wildly excited and there would
be big, red spots in both his cheeks; and when the excitement of the
trip was over he would grow weary. He would cough and want to eat
less and sleep less, but always he was cheerful and a-planning for
railroading with his Dad."
It came time to camp for the night and Jim stopped the story, as he
started our fire and I began to put up our tent.
 CHAPTER XII
THROUGH THE VALLEY WITH THE LITTLE FELLOW

W
hen we had eaten our fill of fried salmon and blue-berry
duff, that no one could stir up and bake better than Jim,
and the camp was tidied for the night, Jim went on with his
story.
He had come to the hard part of the story, the saddest part of his
life, and I was glad that it was dark; I knew it would be easier for
him. I was glad, too, that the camp fire was dying down, for thus I
would see less of suffering that might be revealed could I see his
face in the brighter light.
"I had the Grenfell doctor come. I'd sent ahead to have him met at
the Hospital Ship and a doctor, a great man from the States, on his
vacation, they said, come over here to our place. He was giving his
vacation because he loved Grenfell and the fishermen.
"Little Peter answered all of his questions and I was sheer proud of
him. I could see the Doctor liked the little man. He said to Peter,
when he had finished examining him:
"'I'll make you better, my little man, if I can. You take all the eggs
and milk the hens and the cows will let you have and grow so fat
your mother won't know you.'
"But to me, he said, when he walked down the road a piece with
me:
"'You're Harbor Jim, they tell me, a man loved hereabouts for the
fights you've made to reach the harbor in a night of storm. I am
hating to tell you, Jim, but it's goin' to be a hard fight this time, the
hardest fight you ever had. There's a chance; but one lung is all
gone and the other's bad. I'll do my best, but if you have to go thru
the valley with the little fellow, I'll only hope you won't forget to live
up to your reputation.'
"Then he left me all manner of directions, about eggs and milk that
was to give him ammunition for the fight. Told me to soak him in
sunshine and so on. And I did just as he told me. I gave him his
cod-liver oil, when I had to invent fairy stories to get him to swallow
it. I wrapped him up in blankets and sat him in the sunshine. His
mother did as much or more'n I did. I used to listen of a night to see
if he breathed all right. I listened, when ever Effie was asleep, to see
if I could tell if he breathed as strong as he did the night before.
"Those days my heart was sore all the while, but I couldn't let on for
fear she'd know just how I felt."
Jim swallowed hard, but he had made up his mind to tell me the
story of little Peter and he wasn't the man to back down. He had a
knife and a piece of a birch and he was whittling away. The light
would flare up a moment and I could see him looking straight ahead
into the fire and whittling faster.
"Then I had to cover it up from him; for little Peter was sure that he
was getting better. Seems though the worse he got, the surer he
was he'd be better tomorrow. When he got so weak I had to carry
him across the room, he began to talk more about spring and
railroading again with his dad.
"Sometimes when I'd been off and was comin' home, I dreaded so
seeing him, thus weak, that I'd rather a-gone thru the Narrows on
the darkest night God ever made, than to face Peter with a jolly
quip. So many times then, and so many times since, I have thought,
if I only could have toted the load for him. If only my hand could a-
held it up for him. He was so little and frail and I was big and strong.
And it was the utter, awful helplessness of it that made it so hard to
bear. We wanted to help so bad and there was so very little that
either of us could do.
"We didn't have Clara then. She didn't come until afterward, and
then Peter was all we had. It didn't seem that we could give him up.
I reasoned with myself and I didn't one night forget the Book. But
there were nights when we halted at the springs that our mouths
were so dry and parched that even the Water of Life seemed not to
be sufficient to quench them.
"We went deeper and deeper into the valley. He grew weaker and
weaker. Just like a little flower that is fading away. One night he
grew worse. It was February, and I put on my snow shoes and
started for St. John's for a doctor. I walked away into the night and I
got a doctor and was back afore dawn.
"The doctor took his pulse and said:
"'He'll be crossing at the dawn.'
"Little Peter often listened to the Book and he was beginning to love
it, too; and just before the sun broke that cold, February morning,
he whispered:
"'God is light; in Him is no darkness at all.'
"Then it was morning, but oh, it was night and the valley for us! The
doctor left us and we sat alone, her hand in mine. Effie didn't say
anything; I think if she had I couldn't a bore it. And there was no
minister present. I was glad of that, too. I guess they all want to
help, but a good many on'em that I have knowed want to argue and
to tell you it's all right and you don't want to talk just then and
arguments don't offer much comfort. The time had come when only
one could comfort us and we had to find Him. Some do not find Him
for days, some for weeks, some never find Him again.
"The words that kept saying themselves over to me were these: 'Be
still and know that I am God.' I was some impatient, some bitter. I
know I oughtn't to have been, but I was, and I answered the Lord:
'I am still; see me suffering here; is that all the message?'
"It was a good thing we had something to do. We had to see to the
little wasted body. We had to arrange for the service. We had to tidy
up the house. We shared it all, the new sorrow and the pain, just as
we had shared the wallet and the joys. The minister come way from
St. John's and I was grateful to him. I don't remember just what he
said, but I am sure that Peter was worthy all the good he said of
him; and I know that I needed all the prayer he made.
"But when it was all over and the house was so quiet, it was harder
still. It didn't do no good to listen for his breathing. There was no
need to think of eggs or milk for the little fellow's breakfast. He was
gone!
"I was very tired and I was about to turn in that night after the
funeral, when Effie said:
"'We need to halt by the springs more than ever.'
"I knew she was right, so with a sad heart I opened the Book. I
never knowed just how it was, perhaps the Lord himself guided my
hands, but we come to a little halt at the 14th chapter of John. It
was the Spring of Comfort and Peace, we so much needed. It was
the place where so many have camped before in their night of
sorrow and gone forth strengthened and rejoicing in the morning.
We were very thirsty and it was real water, the water of life and we
drank as we never had drank before. He spoke to us and said: 'Let
not your heart be troubled!'
"I won't repeat that chapter, but it has never lost its power, to
refresh and comfort since the day He first uttered the words. If you
ever have to go again thru the valley yourself; halt there. It will be
the wisest thing you ever did.
"After that I was able to think clear again. I said to myself. I trusted
the Father before and He never did me wrong. I can't just see, but I
can trust and it will grow brighter and so it has, though sometimes I
don't see quite plain, even yet. I know that He must have a place for
the little fellow and He must know what Peter needs. He'll know how
to pick the best teachers and all the experiences he needs. My
Father is looking out for him. He can do no wrong."
For a little while all was quiet but for the chattering of the river as it
hurried on down to sea. The wind freshened in the trees. Messages
were passing above us. Jim brought a bundle of fresh wood and the
fire leaped into a cheerful blaze. There was not any more that
needed to be said. We both made an effort to shake off the sadness
and fell to talking of the weight of the day's salmon catch, as we
undressed. We carried but one little tent and slept together. Some
hour after we had gone to bed, I imagined Jim was trying to find out
if I was asleep without disturbing me. At last he decided that I was
awake and said:
"I'm sure it's all right about little Peter. We're out of the valley now
and are finding again the sunny plain."
 CHAPTER XIII
THE QUEER ONE

"S
artin sure! By the big dipper, it's sartin shame!" Bob
McCartney stood at my door all excitement as he delivered
himself of these explosives.
Bob is a short man and middle-wide, and he is on the increase. This
particular morning he stood on my stoop, the very personification of
heat. He took off his hat and mopped his head and his red face and
without waiting went on with his message.
"The Missus Jim is took sudden and terrible sick. Doc Withers is
there and don't know what ails her. Think of anything she could
take? Anything you know of she could do? Everybody is suggestin'!
Neighbors comin' an' goin' all the while, tryin' to do something for
the Missus Jim. Didn't seem to be anything more I could do. You
can't try everything to onct, so think's I, I'll go and see him. He
comes from New York an' mebbe he'll have a new idea."
"It might be a good thing to let one or two ideas have a chance," I
replied. "I've noticed that ideas that get rushed and crowded don't
do as well."
Bob brightened and pulling on his cap, backed down the stairs. "I'll
tell 'em to go slow and let the first ideas have a chance."
I wisely concluded that Jim would have all the help and more than
he needed and I did not call for three days. When I did Mrs. Jim
herself answered my knock and from just behind Jim shouted:
"She's all right again. Didn't prove so bad as we thought. Something
got inside of her that didn't belong there and soon's it got out, she
come along all right."
"Was it the doctor or you, Jim, that cured her?" I asked, as I sat
down.
"I've been thinking o' that a good deal, this day," he answered.
"Everything traces back to the Almighty, when you let your thought
travel far enough, and I'd like to thank Him, first. I prayed a good
deal and though I don't need no thanks, I believe those prayers
helped. Then the neighbors helped. They loaned hot water bags and
fetched pillows, an' done all manner o' things, 'till thinks I, nobody
ever had such neighbors as us. Then there was Doc Withers. Now
some folks give all the credit to the docs, but I don't; neither do I
take all the praise from 'em. Their His servants, too, and I callate
dividing up the responsibility and the thanks for a cure is a mighty
difficult task. I know I ain't worthy to do it myself."
A knock, a quick, nervous knock came just then and Jim answered
it, throwing wide the door, as he always did, with his cheery, "Come
right in."
A thin, tall man with a long rain-coat and big, black-rimmed glasses
stepped in. Snatching off his gray Alpine hat, he introduced himself.
"I'm Clarence O. Jewett, of Boston. Am visiting in Newfoundland,
spending two and a half days here. Came in on the steamer
'Rosalind' from Halifax, yesterday, going back tomorrow. In St.
John's I was told of Harbor Jim and that his wife was very ill, and I
hired a car and came out here and I am ready to give your wife a
treatment. I have been thinking that perhaps the Lord is using me to
bring the only, real, true religion to Newfoundland. When your wife
has seen the light and comes to know the truth that sin and
everything material is a delusion, deception and a snare, she will
understand that being perfect she cannot really suffer from an
illusion. This earth and all things upon which we look are but
shadows. When your wife is whole again and understands the non-
reality of matter, she will testify and others may hear and heed, until
many on this island will come to praise the Lord and to remember
Clarence O. Jewett, of Boston, who brought the only, real, true
religion—"
At this moment, Mrs. Jim, who had stepped out at the knock, re-
entered the room and Jim had his first chance to speak.
"This is the Missus. The news you received is a little late, for she has
recovered. Since you are a mound-tripper and doin' the country,
probably we ought not to keep you. The road across is about five
hundred miles, and if you're goin' to see any more'n St. John's, you'll
have to hurry afore your ship sails. There was a man down here last
year who staid two days in St. John's and then wrote a book about
Newfoundland, but he skipped a few things."
The man was keenly disappointed. He changed his weight from one
foot to the other, for he had not yet taken the seat that Jim had
offered him. He took off his glasses and wiped them and then
seating himself and clearing his throat, resumed.
"The cure is but temporary. Your wife will not be well until she has
learned that there is but one thing to know and that is the truth and
the truth about the truth. And though you cannot expect to
understand it, I will start you on the way toward the one, only, real,
true religion."
"Am I supposed just to listen?" asked Jim, "or do you think I might
know enough to ask a question now and then?"
"Certainly, certainly," the queer man replied. "I have an answer for
every question that is absolutely logical. Take the question of the
existence of evil; that is the most puzzling question in all the world. I
have an answer to it that is entirely satisfactory. Nobody can
contradict it. Evil is matter. Matter does not exist. Therefore evil does
not exist and since it does not exist, it never could have been
created. Evil and matter are just wrong statements of mind. Do you
see? Is it perfectly clear to you?"
Jim gulped, as though he was in swimming and had accidentally
swallowed some salt water. I had come to have a profound
admiration for Jim and was coming more and more to appreciate his
wholesome philosophy, and now I was waiting to see what Jim
would do with this man's statements.
"You have doused me beyond my depth, I guess," was the
somewhat puzzled reply of Jim. "It isn't plain to me. But heave
ahead a little and mebbe I'll get some idea of what port you're sailin'
to. The only thing you have said so far that has any familiar sound to
me is what you said about the one, only, real, true religion. I've
heard that several times before. Seems though most every kind o'
religion and every different church feels that it's got the one, only,
real, true religion. Strikes me, every blessed one on'em has got
some of the real religion and also some foolishness and smallness
and no one on'em has got the pure, undiluted article that Jesus
Christ brought to the world. I think He come the nearest to livin' the
real religion. But how'd you discover that your's was the only
religion?"
The queer man evidently thought the question irrelevant, for he was
off again.
"It is now proved that all is mental or mind. Your thoughts are the
opposites of mind. They do not exist. They are even as all other
things, non-existent, non-real. God is the only reality. There is no
thing outside of God. You are not separated from Him."
"Then," interrupted Jim, "how about the Prodigal Son? Didn't he get
separated from his Father?"
"That is speaking in terms of no-mind. You have not yet grasped the
thought. Nothing can exist but good. God never saw the Prodigal
Son until he came back, because he never has or can see anything
evil."
"Your God may not see or know evil, sickness or suffering or
anything that makes a man miserable. I say, your God mayn't, but
mine does. It's his knowledge that makes Him compassionate. If He
didn't know what was happening to His own children, that He had
created and planned for, then I'd rather pray to Bob McCartney.
Think, sir, what kind of a mother would your mother a-been, if she
hadn't known when you cried, and you hadn't a-been able to climb
up and lay in her arms and be comforted and forgiven? She wouldn't
a-been a mother and God wouldn't be a God unless He knew what
was a-happening to His own children! Why man alive didn't He make
the world; aren't they His, the cattle on a thousand hills, the
lightenin' and the thunder, the sweet grass and the flowers and all
things in and on and under the earth? If He has gone off and
forgotten it all and don't know good and evil, if He don't know when
we're tired and failin' and tryin' again, why what would be the use o'
prayer or, for that matter, for livin' at all?"
The queer man, at this point, removed his rubbers, but made no
comment upon Jim's questions. Perhaps his feet were so warm it
was hard for him to keep his head cool.
"You are utterly deceived," he continued. "You are confusing the real
and the non-real. You are following after shadows that do not exist
at all. You do not know the truth. You are bound. You are looking at
the mist of matter that will disappear as the knowledge of truth
develops within you. If you will begin to deny the existence of evil,
you will begin to banish disease and ultimately you will understand
that all things are but illusions."
"Pears to me," Jim said, as the queer man paused for breath, before
launching more sentences about the truth. "Pears to me, you're
sailin' round in a circle, and havin' a hard time dodging the winds o'
logic. Call the flower, the mountain, and the man, shadows and
illusions; if you will. I don't object to that, only I want you to agree
with me that they are beautiful. The only thing I am afeared of is
that you'll make some folks think this is not His world at all; and I
want them to know that this is His world and that He planned these
things you have re-named shadows and illusions. I callate there's
danger in your statements when you come to follow them out. Then,
too, these shadows have been actin' about uniform for as long ago
as the book o' Genesis and afore that, and I don't propose to try to
get much farther back, for it makes my eyes ache to see back o'
that.
"When you tell me this body o' mine is an illusion, it kinder riles me,
for I believe the Good Father planned this body as much as He
planned a soul for me. It's a house for my soul as long as I'm in this
earth and I callate it's to be treated holy while it houses my soul. I
know it will get kinder old and dingy bye and bye and I'll be quitting
it, but that ain't no good reason for neglectin' it now.
"Of course if what you say was true and there was no material and it
was all in thinking, then we wouldn't have to wear clothes, nor eat
food and you wouldn't have to wear your specs, nor your goloshes,
because it's a little damp under feet this morning. You may be
different, Mr. Jewett, with your one, only, real, true religion, but we
Landers up here all get a little older as days go by; we all like to be
cheered by food and fuel, and we all feel the difference between
winter and summer, and we all travel sooner or later to the better
land. Seems to be His plan."
The queer man was gathering words for new statements; but while
he was listening to the last of Jim's replies, he was looking intently
at his hands. If it may be permitted to speak in ordinary fashion of a
man of his philosophy, his hands were dirty and he had become
painfully aware of it. Jim noticed his concern and remarked with a
certain acerbity of tone:
"You don't clean your hands with soap and water, do you?"
The queer man in turn showed some increase of warmth as he
replied:
"I certainly do when I need to, that's only common sense."
"Well," mused Jim, this time very slowly, "do you know, I don't
believe in using too much soap, it's caustic and it's harmful
sometimes to the skin, but do you know, once in a while I get a bit
riled and dirty inside o' me and I decide that it's only common sense
to clean that just as I would my hands."
The queer man sniffed and asked for a Bible. "Have you a Bible?"
Welcome to Our Bookstore - The Ultimate Destination for Book Lovers
Are you passionate about testbank and eager to explore new worlds of
knowledge? At our website, we offer a vast collection of books that
cater to every interest and age group. From classic literature to
specialized publications, self-help books, and children’s stories, we
have it all! Each book is a gateway to new adventures, helping you
expand your knowledge and nourish your soul
Experience Convenient and Enjoyable Book Shopping Our website is more
than just an online bookstore—it’s a bridge connecting readers to the
timeless values of culture and wisdom. With a sleek and user-friendly
interface and a smart search system, you can find your favorite books
quickly and easily. Enjoy special promotions, fast home delivery, and
a seamless shopping experience that saves you time and enhances your
love for reading.
Let us accompany you on the journey of exploring knowledge and
personal growth!

ebooksecure.com