Find the Full Original Textbook (PDF) in the link

below:
CLICK HERE
Detailed Contents
Preface xvi
1 Introduction to Business Analytics 1
1-1 Introduction 3
1-2 Overview of the Book 4
1-2a The Methods 4
1-2b The Software 6
1-3 Introduction to Spreadsheet Modeling 8
1-3a Basic Spreadsheet Modeling: Concepts and
Best Practices 9
1-3b Cost Projections 12
1-3c Breakeven Analysis 15
1-3d Ordering with Quantity Discounts and Demand
Uncertainty 20
1-3e Estimating the Relationship between Price and
Demand 24
1-3f Decisions Involving the Time Value of Money
29
1-4 Conclusion 33
PART 1 Data Analysis 37
2 Describing the Distribution of a Variable 38
2-1 Introduction 39
2-2 Basic Concepts 41
2-2a Populations and Samples 41
2-2b Data Sets, Variables, and Observations 41
2-2c Data Types 42
2-3 Summarizing Categorical Variables 45
2-4 Summarizing Numeric Variables 49
2-4a Numeric Summary Measures 49
2-4b Charts for Numeric Variables 57
2-5 Time Series Data 62
2-6 Outliers and Missing Values 69
2-7 Excel Tables for Filtering, Sorting, and
Summarizing 71
2-8 Conclusion 77
Appendix: Introduction to StatTools 83
3 Finding Relationships among Variables 84
3-1 Introduction 85
3-2 Relationships among Categorical Variables 86
3-3 Relationships among Categorical Variables
and a Numeric Variable 89
3-4 Relationships among Numeric Variables 96
3-4a Scatterplots 96
3-4b Correlation and Covariance 101
3-5 Pivot Tables 106
3-6 Conclusion 126
Appendix: Using StatTools to Find Relationships 131
4 Business Intelligence (BI) Tools for Data Analysis
132
4-1 Introduction 133
4-2 Importing Data into Excel with Power Query 134
4-2a Introduction to Relational Databases 134
4-2b Excel’s Data Model 139
4-2c Creating and Editing Queries 146
4-3 Data Analysis with Power Pivot 152
4-3a Basing Pivot Tables on a Data Model 154
4-3b Calculated Columns, Measures, and the DAX
Language 154
4-4 Data Visualization with Tableau Public 162
4-5 Data Cleansing 172
4-6 Conclusion 178
PART 2 Probability and Decision Making under
Uncertainty 183
5 Probability and Probability Distributions 184
5-1 Introduction 185
5-2 Probability Essentials 186
5-2a Rule of Complements 187
5-2b Addition Rule 187
5-2c Conditional Probability and the Multiplication
Rule 188
5-2d Probabilistic Independence 190
5-2e Equally Likely Events 191
5-2f Subjective Versus Objective Probabilities 192
5-3 Probability Distribution of a Random Variable
194
5-3a Summary Measures of a Probability Distribution
195
5-3b Conditional Mean and Variance 198
5-4 The Normal Distribution 200
5-4a Continuous Distributions and Density Functions
200
5-4b The Normal Density Function 201
5-4c Standardizing: Z-Values 202
5-4d Normal Tables and Z-Values 204
5-4e Normal Calculations in Excel 205
5-4f Empirical Rules Revisited 208
5-4g Weighted Sums of Normal Random Variables
208
5-4h Normal Distribution Examples 209
5-5 The Binomial Distribution 214
5-5a Mean and Standard Deviation of the Binomial
Distribution 217
5-5b The Binomial Distribution in the Context of
Sampling 217
5-5c The Normal Approximation to the Binomial 218
5-5d Binomial Distribution Examples 219
5-6 The Poisson and Exponential Distributions 226
5-6a The Poisson Distribution 227
5-6b The Exponential Distribution 229
5-7 Conclusion 231
6 Decision Making under Uncertainty 242
6-1 Introduction 243
6-2 Elements of Decision Analysis 244
6-3 EMV and Decision Trees 247
6-4 One-Stage Decision Problems 251
6-5 The PrecisionTree Add-In 254
6-6 Multistage Decision Problems 257
6.6a Bayes’ Rule 262
6-6b The Value of Information 267
6-6c Sensitivity Analysis 270
6-7 The Role of Risk Aversion 274
6-7a Utility Functions 275
6-7b Exponential Utility 275
6-7c Certainty Equivalents 278
6-7d Is Expected Utility Maximization Used? 279
6-8 Conclusion 280
PART 3 Statistical Inference 293
7 Sampling and Sampling Distributions 294
7-1 Introduction 295
7-2 Sampling Terminology 295
7-3 Methods for Selecting Random Samples 297
7-3a Simple Random Sampling 297
7-3b Systematic Sampling 301
7-3c Stratified Sampling 301
7-3d Cluster Sampling 303
7-3e Multistage Sampling 303
7-4 Introduction to Estimation 305
7-4a Sources of Estimation Error 305
7-4b Key Terms in Sampling 306
7-4c Sampling Distribution of the Sample Mean 307
7-4d The Central Limit Theorem 312
7-4e Sample Size Selection 317
7-4f Summary of Key Ideas in Simple Random
Sampling 318
7-5 Conclusion 320
8 Confidence Interval Estimation 323
8-1 Introduction 323
8-2 Sampling Distributions 325
8-2a The t Distribution 326
8-2b Other Sampling Distributions 327
8-3 Confidence Interval for a Mean 328
8-4 Confidence Interval for a Total 333
8-5 Confidence Interval for a Proportion 336
8-6 Confidence Interval for a Standard Deviation 340
8-7 Confidence Interval for the Difference between
Means 343
8-7a Independent Samples 344
8-7b Paired Samples 346
8-8 Confidence Interval for the Difference between
Proportions 348
8-9 Sample Size Selection 351
8-10 Conclusion 358
9 Hypothesis Testing 368
9-1 Introduction 369
9-2 Concepts in Hypothesis Testing 370
9-2a Null and Alternative Hypotheses 370
9-2b One-Tailed Versus Two-Tailed Tests 371
9-2c Types of Errors 372
9-2d Significance Level and Rejection Region 372
9-2e Significance from p-values 373
9-2f Type II Errors and Power 375
9-2g Hypothesis Tests and Confidence Intervals 375
9-2h Practical Versus Statistical Significance 375
9-3 Hypothesis Tests for a Population Mean 376
9-4 Hypothesis Tests for Other Parameters 380
9-4a Hypothesis Test for a Population Proportion
380
9-4b Hypothesis Tests for Difference between
Population Means 382
9-4c Hypothesis Test for Equal Population
Variances 388
9-4d Hypothesis Test for Difference between
Population Proportions 388
9-5 Tests for Normality 395
9-6 Chi-Square Test for Independence 401
9-7 Conclusion 404
PART 4 Regression Analysis and Time Series
Forecasting 411
10 Regression Analysis: Estimating Relationships
412
10-1 Introduction 413
10-2 Scatterplots: Graphing Relationships 415
10-3 Correlations: Indicators of Linear Relationships
422
10-4 Simple Linear Regression 424
10-4a Least Squares Estimation 424
10-4b Standard Error of Estimate 431
10-4c R-Square 432
10-5 Multiple Regression 435
10-5a Interpretation of Regression Coefficients 436
10-5b Interpretation of Standard Error of Estimate
and R-Square 439
10-6 Modeling Possibilities 442
10-6a Dummy Variables 442
10-6b Interaction Variables 448
10-6c Nonlinear Transformations 452
10-7 Validation of the Fit 461
10-8 Conclusion 463
11 Regression Analysis: Statistical Inference 472
11-1 Introduction 473
11-2 The Statistical Model 474
11-3 Inferences About the Regression Coefficients
477
11-3a Sampling Distribution of the Regression
Coefficients 478
11-3b Hypothesis Tests for the Regression
Coefficients and p-Values 480
11-3c A Test for the Overall Fit: The ANOVA Table
481
11-4 Multicollinearity 485
11-5 Include/Exclude Decisions 489
11-6 Stepwise Regression 494
11-7 Outliers 499
11-8 Violations of Regression Assumptions 504
11-8a Nonconstant Error Variance 504
11-8b Nonnormality of Residuals 504
11-8c Autocorrelated Residuals 505
11-9 Prediction 507
11-10 Conclusion 512
12 Time Series Analysis and Forecasting 523
12-1 Introduction 524
12-2 Forecasting Methods: An Overview 525
12-2a Extrapolation Models 525
12-2b Econometric Models 526
12-2c Combining Forecasts 526
12-2d Components of Time Series Data 527
12-2e Measures of Accuracy 529
12-3 Testing for Randomness 531
12-3a The Runs Test 534
12-3b Autocorrelation 535
12-4 Regression-Based Trend Models 539
12-4a Linear Trend 539
12-4b Exponential Trend 541
12-5 The Random Walk Model 544
12-6 Moving Averages Forecasts 547
12-7 Exponential Smoothing Forecasts 551
12-7a Simple Exponential Smoothing 552
12-7b Holt’s Model for Trend 556
12-8 Seasonal Models 560
12-8a Winters’ Exponential Smoothing Model 561
12-8b Deseasonalizing: The Ratio-to-Moving-
Averages Method 564
12-8c Estimating Seasonality with Regression 565
12-9 Conclusion 569
PART 5 Optimization and Simulation Modeling 575
13 Introduction to Optimization Modeling 576
13-1 Introduction 577
13-2 Introduction to Optimization 577
13-3 A Two-Variable Product Mix Model 579
13-4 Sensitivity Analysis 590
13-4a Solver’s Sensitivity Report 590
13-4b SolverTable Add-In 593
13-4c A Comparison of Solver’s Sensitivity Report
and SolverTable 599
13-5 Properties of Linear Models 600
13-6 Infeasibility and Unboundedness 602
13-7 A Larger Product Mix Model 604
13-8 A Multiperiod Production Model 612
13-9 A Comparison of Algebraic and Spreadsheet
Models 619
13-10 A Decision Support System 620
13-11 Conclusion 622
14 Optimization Models 630
14-1 Introduction 631
14-2 Employee Scheduling Models 632
14-3 Blending Models 638
14-4 Logistics Models 644
14-4a Transportation Models 644
14-4b More General Logistics Models 651
14-5 Aggregate Planning Models 659
14-6 Financial Models 667
14-7 Integer Optimization Models 677
14-7a Capital Budgeting Models 678
14-7b Fixed-Cost Models 682
14-7c Set-Covering Models 689
14-8 Nonlinear Optimization Models 695
14-8a Difficult Issues in Nonlinear Optimization 695
14-8b Managerial Economics Models 696
14-8c Portfolio Optimization Models 700
14-9 Conclusion 708
15 Introduction to Simulation Modeling 717
15-1 Introduction 718
15-2 Probability Distributions for Input Variables 720
15-2a Types of Probability Distributions 721
15-2b Common Probability Distributions 724
15-2c Using @RISK to Explore Probability
Distributions 728
15-3 Simulation and the Flaw of Averages 736
15-4 Simulation with Built-in Excel Tools 738
15-5 Simulation with @RISK 747
15-5a @RISK Features 748
15-5b Loading @RISK 748
15-5c @RISK Models with a Single Random Input
749
15-5d Some Limitations of @RISK 758
15-5e @RISK Models with Several Random Inputs
758
15-6 The Effects of Input Distributions on Results
763
15-6a Effect of the Shape of the Input Distribution(s)
763
15-6b Effect of Correlated Inputs 766
15-7 Conclusion 771
16 Simulation Models 779
16-1 Introduction 780
16-2 Operations Models 780
16-2a Bidding for Contracts 780
16-2b Warranty Costs 784
16-2c Drug Production with Uncertain Yield 789
16-3 Financial Models 794
16-3a Financial Planning Models 795
16-3b Cash Balance Models 799
16-3c Investment Models 803
16-4 Marketing Models 810
16-4a Customer Loyalty Models 810
16-4b Marketing and Sales Models 817
16-5 Simulating Games of Chance 823
16-5a Simulating the Game of Craps 823
16-5b Simulating the NCAA Basketball Tournament
825
16-6 Conclusion 828
PART 6 Advanced Data Analysis 837
17 Data Mining 838
17-1 Introduction 839
17-2 Classification Methods 840
17-2a Logistic Regression 841
17-2b Neural Networks 846
17-2c Naïve Bayes 851
17-2d Classification Trees 854
17-2e Measures of Classification Accuracy 855
17-2f Classification with Rare Events 857
17-3 Clustering Methods 860
17-4 Conclusion 870
18 Analysis of Variance and Experimental Design
(MindTap Reader only)
18-1 Introduction 18-2
18-2 One-Way ANOVA 18-5
18-2a The Equal-Means Test 18-5
18-2b Confidence Intervals for Differences Between
Means 18-7
18-2c Using a Logarithmic Transformation 18-11
18-3 Using Regression to Perform ANOVA 18-15
18-4 The Multiple Comparison Problem 18-18
18-5 Two-Way ANOVA 18-22
18-5a Confidence Intervals for Contrasts 18-28
18-5b Assumptions of Two-Way ANOVA 18-30
18-6 More About Experimental Design 18-32
18-6a Randomization 18-32
18-6b Blocking 18-35
18-6c Incomplete Designs 18-38
18-7 Conclusion 18-40
19 Statistical Process Control (MindTap Reader
only)
19-1 Introduction 19-2
19-2 Deming’s 14 Points 19-3
19-3 Introduction to Control Charts 19-6
19-4 Control Charts for Variables 19-8
19-4a Control Charts and Hypothesis Testing 19-13
19-4b Other Out-of-Control Indications 19-15
19-4c Rational Subsamples 19-16
19-4d Deming’s Funnel Experiment and Tampering
19-18
19-4e Control Charts in the Service Industry 19-22
19-5 Control Charts for Attributes 19-26
19-5a P Charts 19-26
19-5b Deming’s Red Bead Experiment 19-29
19-6 Process Capability 19-33
19-6a Process Capability Indexes 19-35
19-6b More on Motorola and 6-Sigma 19-40
19-7 Conclusion 19-43
APPENDIX A: Quantitative Reporting (MindTap
Reader only)
A-1 Introduction A-1
A-2 Suggestions for Good Quantitative Reporting A-
2
A-2a Planning A-2
A-2b Developing a Report A-3
A-2c Be Clear A-4
A-2d Be Concise A-4
A-2e Be Precise A-5
A-3 Examples of Quantitative Reports A-6
A-4 Conclusion A-16
References 873
Index 875

CHAPTER 1 Introduction to Business Analytics

BUSINESS ANALYTICS PROVIDES INSIGHTS
AND IMPROVES PERFORMANCE

This book is about analyzing data and using

quantitative modeling to help companies understand
their business, make better decisions, and improve
performance. We have been teaching the methods
discussed in this book for decades, and companies
have been using these methods for decades to
improve performance and save millions of dollars.
There- fore, we were a bit surprised when a brand
new term, Busi- ness Analytics (BA), became hugely
popular several years ago. All of a sudden, BA
promised to be the road to success. By using
quantitative BA methods—data analysis, optimiza-
tion, simulation, prediction, and others—companies
could drastically improve business performance.
Haven’t those of us in our field been doing this for
years? What is different about BA that has made it
so popular, both in the academic world and even
more so in the business world? The truth is that BA
does use the same quantitative methods that have
been used for years, the same methods you will
learn in this book. BA has not all of a sudden
invented brand new quantitative methods to eclipse
traditional methods. The main difference is that BA
uses big data to solve business problems and
provide insights. Companies now have access to
huge sources of data, and the technology is now
available to use huge data sets for quantitative
analysis, predictive modeling, optimization, and
simulation. In short, the same quantitative methods
that have been used for years can now be even
more effective by utilizing big data and the
corresponding technology. For a quick introduction
to BA, you should visit the BA Wikipedia site (search
the Web for “business analytics”). Among other
things, it lists areas where BA plays a prom- inent
role, including the following: retail sales analytics;
financial services analytics; risk and credit analytics;
marketing analytics; pricing analytics; supply chain
analytics; and transportation analytics. If you glance
through the examples and problems in this book,
you will see that most of them come from these
same areas. Again, the difference is that we use
relatively small data sets to get you started—we do
not want to overwhelm you with gigabytes of data—
whereas real applications of BA use huge data sets
to advantage. A more extensive discussion of BA
can be found in the Fall 2011 research report,
Analytics: The Widening Divide, published in the MIT
Sloan Management Review in collaboration with
IBM, a key developer of BA software (search the
Web for the arti- cle’s title). This 22-page article
discusses what BA is and provides several case
studies. In addition, it lists three key competencies
people need to compete successfully in the BA
world—and hopefully you will be one of these
people. • Competency 1: Information management
skills to manage the data. This competency involves
expertise in a variety of techniques for managing
data. Given the key role of data in BA methods, data
quality is extremely important. With data coming
from a number of disparate sources, both internal
and external to an organization, achieving data
quality is no small feat.
• Competency 2: Analytics skills and tools to
understand the data. We were not surprised, but
rather very happy, to see this competency listed
among the requirements because these skills are
exactly the skills we cover throughout this book—
optimization with advanced quantitative algorithms,
simulation, and others. • Competency 3: Data-
oriented culture to act on the data. This refers to the
culture within the organization. Everyone involved,
especially top management, must believe strongly in
fact-based decisions arrived at using analytical
methods. The article argues persuasively that the
companies that have these competencies and have
embraced BA have a distinct competitive advantage
over companies that are just start- ing to use BA
methods or are not using them at all. This explains
the title of the article. The gap between companies
that embrace BA and those that do not will only
widen in the future. This field of BA and, more
specifically, data analysis is progressing very
quickly, not only in academic areas but also in many
applied areas. By analyzing big data sets from
various sources, people are learning more and more
about the way the world works. One recent book,
Everybody Lies: Big Data New Data, and What the
Internet Can Tell Us About Who We Really Are, by
Stephens-Davidowitz (2017), is especially
interesting. The title of the book suggests that
people lie to themselves, friends, Face- book, and
surveys, but they don’t lie on Google searches. The
author, a social scientist, has focused on the data
from Google searches to study wide-ranging topics,
including economics, finance, education, sports,
gambling, racism, and sex. Besides his insightful
and often unexpected findings, the author presents
several general conclusions about analysis with big
data: • Not surprisingly, insights are more likely to be
obtained when analyzing data in novel ways, such
as analyzing the organ sizes of race horses instead
of their pedigrees. (A data analyst used data on race
horse heart size to predict that American Pharaoh
would be great. The horse went on to win the Triple
Crown in 2015, the first horse to do so since 1978.)
On the other hand, if all available data have been
analyzed in every conceivable way, as has been
done with stock market data, even analysis with big
data is unlikely to produce breakthroughs. • Huge
data sets allow analysts to zoom in on smaller
geographical areas or smaller subsets of a
population for keener insights. For example, when
analyzing why some areas of the U.S. produce more
people who warrant an entry on Wikipedia, you
might find that richer and more educated states have
more success. However, with a data set of more
than 150,000 Americans of Wikipedia fame, the
author was able to pinpoint that a disproportionate
number were born in college towns. Traditional small
data sets don’t allow drilling down to this level of
detail. • Companies are increasingly able to run
controlled experiments, usually on the Web, to learn
the causes of behavior. The new term for these
experiments is A/B testing. The purpose of a test is
to determine whether a person prefers A or B. For
example, Google routinely runs A/B tests, showing
one randomly selected group of customers one thing
and a second randomly selected group another. The
difference could be as slight as the presence or
absence of an arrow link on a Web page. Google
gets immediate results from such tests by studying
the customers’ clicking behavior. The company can
then use the results of one test to lead to another
follow-up test. A/B testing is evidently happening all
the time, whether or not we’re aware of it. • A
potential drawback of analyzing big data is
dimensionality, where there are so many potential
explanatory variables that one will almost surely look
like a winner—even though it isn’t. The author
illustrates this by imagining daily coin flips of 1000
coins and tracking which of them comes up heads
on days when the stock market is up. One of these
coins, just by chance, is likely to correlate highly with
the stock market, but this coin is hardly a useful
predictor of future stock market movement.

Find the Full Original Textbook (PDF) in the link

below:

CLICK HERE