Analytics Stories: Using Data to Make Good Things Happen

ISBN-13: 9781119646037 / Angielski / Miękka / 2020 / 528 str.

Wayne L. Winston

Spis treści

Introduction xxviiPart I What Happened? 1Chapter 1 Preliminaries 3Basic Concepts in Data Analysis 3What Is a Random Variable? 9Excel Calculations 13Chapter 2 Was the 1969 Draft Lottery Fair? 17The Data 17The Analysis 18Excel Calculations 20Chapter 3 Who Won the 2000 Election: Bush or Gore? 23Projecting the Undervotes 24What Happened with the Overvotes? 25The Butterfl y Did It! 25Excel Calculations 28Chapter 4 Was Liverpool Over Barcelona the Greatest Upset in Sports History? 31How Should We Rank Upsets? 31Leicester Wins the 2015-2016 Premier League 32#16 Seed UMBC Beats #1 Seed Virginia 33The Jets Win Super Bowl III 33Other Big Upsets 34Chapter 5 How Did Bernie Madoff Keep His Fund Going? 35The Mathematics of Ponzi Schemes 36Madoff's Purported Strategy 37The Sharpe Ratio Proves Madoff Was a Fraud 39Benford's Law and Madoff's Fraud 40Excel Calculations 41Chapter 6 Is the Lot of the American Worker Improving? 45Is U.S. Family Income Skewed? 45Median Income and Politics 46Causes of Increasing U.S. Income Inequality 48Money Isn't Everything: The HumanDevelopment Index 50Create Your Own Ranking of Well-Being 50Are Other Countries Catching Up to the U.S.? 51Excel Calculations 52Chapter 7 Measuring Income Inequality with the Gini, Palm, and Atkinson Indices 53The Gini Index 53The Palma Index 56The Atkinson Index 57Excel Calculations 59Chapter 8 Modeling Relationships Between Two Variables 61Examples of Relationships Between Two Variables 61Finding the Best-Fitting (Least Squares) Line 62Computing the Beta of a Stock 63What Is a Good R2? 64Correlation and R2 65We are Not Living in a Linear World 67Excel Calculations 69Chapter 9 Intergenerational Mobility 73Absolute Intergenerational Mobility 74Intergenerational Elasticity 74Rank-Rank Mobility 75Comparing IGE and Rank-Rank Mobility 75Measuring Mobility with Quintiles 78The Great Gatsby Curve 80Excel Calculations 82Chapter 10 Is Anderson Elementary School a Bad School? 85How Can We Adjust for Family Income? 86Estimating the Least Squares Line 86Can We Compare Standardized Test Performance for Students in Different States? 86Excel Calculations 87Chapter 11 Value-Added Assessments of Teacher Effectiveness 89Simple Gain Score Assessment 90Covariate Adjustment Assessment 91Layered Assessment Model 91Cross-Classified Constant Growth Assessment 91Problems with VAA 93How Much Is a Good Teacher Worth? 94Excel Calculations 95Chapter 12 Berkeley, Buses, Cars, and Planes 97Simpson's Paradox and College Admissions 98The Waiting Time Paradox 100When Is the Average of 40 and 80 Not 60? 100Why Pre COVID Were There Never EmptySeats on My Flight? 101Excel Calculations 101Chapter 13 Is Carmelo Anthony a Hall of Famer? 103What Metric Defines Basketball Ability? 104Wins Above Replacement Player (WARP) 105Manu, Melo, Dirk, and Dwayne 106How Do 25,000 Points Lead to So Few Wins? 106Chapter 14 Was Derek Jeter a Great Fielder? 109Fielding Statistics: The First Hundred Years 109Range Factor 110The Fielding Bible: A Great Leap Forward 111The Next Frontier 112Chapter 15 "Drive for Show and Putt for Dough?" 115Strokes Gained 115The Myth Exposed 116Chapter 16 What's Wrong with the NFL QB Rating? 117NFL Quarterback Rating 117ESPN's Total Quarterback Rating 124Excel Calculations 125Chapter 17 Some Sports Have All the Luck 127Skill vs. Luck: The Key Idea 127The Results 129Chapter 18 Gerrymandering 131A Stylized Example 132The Mathematics of Gerrymandering 136Chapter 19 Evidence-Based Medicine 143James Lind and Scurvy: The Birth of Evidence-Based Medicine 143The Randomized Streptomycin Tuberculosis Trial 145Excel Calculations 146Hormone Replacement: Good or Bad? 148Chapter 20 How Do We Compare Hospitals? 151Ratings Criteria 152Conclusion 156Excel Calculations 157Chapter 21 What Is the Worst Health Care Problem in My Country? 159Disability-Adjusted Life Years 159Determination of Disability Weights 160To Age Weight or Discount, That Is the Question 162Key Facts About World Health 163Part II What Will Happen? 167Chapter 22 Does a Mutual Fund's Past Performance Predict Future Performance? 169Mutual Fund Basics 170Morningstar Ratings 170Risk-Adjusting Fund Returns 171How Well Do Morningstar Star RatingsPredict a Fund's Future Performance? 175The Effect of Expense Ratio on Long-Term Performance 177Excel Calculations 178Chapter 23 Is Vegas Good at Picking NFL Games? 181How NFL Betting Works 181Bias and Accuracy 184Vegas Forecasts are Unbiased 185Totals Predictions and Money Line Predictions are Unbiased 188NFL Accuracy: The Line vs. the Computers 188A System Works Until It Doesn't 189Chapter 24 Will My New Hires Be Good Employees? 191What Data Do We Need to Determine Attributes That Best Predict Employee Performance? 192Besides GMA, Not Much Affects Job Performance 196Excel Calculations 197Chapter 25 Should I Go to State U or Princeton? 199Analyzing Princeton vs. Penn State 200Excel Calculations 202Chapter 26 Will My Favorite Sports Team Be Great Next Year? 203Francis Galton and Regression to the Mean 203Regression to the Mean in the NFL and the NBA 204Excel Calculations 207Chapter 27 How Did Central Bankers Fail to Predict the 2008 Recession? 209The Inverted Yield Curve 210The Sahm Rule: Early Warning Signal for Recession 211Control Charts and the Housing Price/Rent Ratio 211Excel Calculations 215Chapter 28 How Does Target Know If You're Pregnant? 219What Available Data Can Be Usedto Identify Pregnant Women? 220Problems Arise 220An Example of a Pregnancy Prediction Score 221Chapter 29 How Does Netflix Recommend Movies and TV Shows? 225User-Based Collaborative Filtering 226Item-Based Filtering 229Chapter 30 Can We Predict Heart Attacks in Real Time? 233Posterior Probability 234Sensitivity and Specifi city 235ROC Curve 235Back to the Apple Heart Study 237AliveCor and KardiaBand 239Chapter 31 Is Proactive Policing Effective? 241Hot Spots Policing 242Predictive Policing 243CCTV 244Stop and Frisk 244Broken Windows 246Excel Calculations 247Chapter 32 Guess How Many are Coming to Dinner? 249Which Parameters Must Be Estimated? 250The Data 252The Results 253Which Factor Really Matters? 254Excel Calculations 254Chapter 33 Can Prediction Markets Predict the Future? 259Examples of Trade Contracts 260Prediction Market Trading Mechanisms 261Accuracy of Prediction Markets and Wisdom of Crowds 262Chapter 34 The ABCs of Polling 265Why are 1,112 People Enough to Represent U.S. Voters? 265Why Doesn't a Larger Population Require a Larger Sample Size? 267So, What Can Go Wrong? 268Rating Polls 271Chapter 35 How Did Buzzfeed Make the Dress Go Viral? 273Measuring Instagram Engagement 274Tweets Do Not Always Go Viral Immediately 274Do the First Few Days Predict the Future of a Meme? 275Chapter 36 Predicting Game of Thrones TV Ratings 277What Does Google Trends Tell Us? 277Predicting the Present with Google Trends 278Using Google Trends to Forecast GOT Ratings 279Excel Calculations 281Part III Why Did It Happened? 283Chapter 37 Does Smoking Cause Lung Cancer? 285Correlation and Causation Redux 285The Key Evidence 286Could Air Pollution Have Caused Lung Cancer? 287The Cigarette Companies Hit Back 287Excel Calculations 288Chapter 38 Why are the Houston Rockets a Good Basketball Team? 291NBA Shooting Math 101 292Zach LaVine Battles the Bulls' Analytics Department 295Conclusion 296Excel Calculations 296Chapter 39 Why Have Sacrifice Bunts and Intentional Walks Nearly Disappeared? 297The Case Against Bunting 298Bunting Against the Shift 299Why are Intentional Walks on the Decline? 300Chapter 40 Do NFL Teams Pass Too Much and Go for It Often Enough on Fourth Down? 301The Ascent of Passing 301Fourth Down Strategy 303New Data Partially Vindicates the Coaches 304Teams Should Go for Two More Often 306Chapter 41 What Caused the 1854 London Cholera Outbreak? 307Cholera 307Snow and the Broad Street Pump 308Snow's Randomized Controlled Trial 310Conclusion 311Excel Calculations 312Chapter 42 What Affects the Sales of a Retail Product? 313Painter's Tape 313Estimating the Model Parameters 315Excel Calculations 316Chapter 43 Why Does the Pareto Principle Explain So Many Things? 319Power Laws 320Why Do Incomes Follow the Pareto Principle? 322Why Do a Few Websites Get Most of the Hits? 323Excel Calculations 324Chapter 44 Does Where You Grow Up Matter? 327Quasi-Experimental Design vs. Randomized Controlled Trials 328What Drives Neighborhood Differences in Upward Mobility? 329How Can We Make Things Better? 330Chapter 45 The Waiting is the Hardest Part 333Which Factors Influence the Performance of a Queueing System? 334Operating Characteristics of a Queueing System 334How Does Variability Degrade the Performance of a Queueing System? 335Calculating the Operating Characteristics of a Queueing System 336Excel Calculations 338Chapter 46 are Roundabouts a Good Idea? 339What Is a Roundabout? 340History of Roundabouts 340Benefi ts of Roundabouts 341Disadvantages of Roundabouts 343Roundabout Capacity 344Roundabouts and Revolutions 345Chapter 47 Red Light, Green Light, or No Light? 347What Causes Traffic Jams? 347How Should We Set the Lights? 348Ramp Meters and Equity 349Measuring the Impact of Ramp Meters 350The Twin Cities Metering Holiday 350Part IV How Do I Make Good Things Happen? 351Chapter 48 How Can We Improve K-12 Education? 353Tennessee's STAR Study on K-2 Class Size 355Cost-Benefi t Analysis 356Can Predictive Analytics Increase Enrollment and Performance in Eighth-Grade Algebra I? 360Excel Calculations 360Chapter 49 Can A/B Testing Improve My Website's Performance? 363Improving Obama's Fundraising in 2008 364The Mechanics of Resampling 365Excel Calculations 366Chapter 50 How Should I Allocate My Retirement Portfolio? 369The Basic Portfolio Optimization Model 369The Effi cient Frontier 372Diffi culties in Implementing the Markowitz Model 374Excel Calculations 374Chapter 51 How Do Hedge Funds Work? 377Growth in Hedge Funds and Hedge Fund Fee Structure 378Shorting a Stock 378Long/Short and Market-Neutral Strategies 378Convertible Arbitrage 380Merger Arbitrage 382Global Macro Strategy 383Hedge Fund Performance 384The George Costanza Portfolio 384Excel Calculations 385Chapter 52 How Much Should We Order and When Should We Order? 389The Economic Order Quantity Model 389Reorder Points, Service Levels, and Safety Stock 392Excel Calculations 393Chapter 53 How Does the UPS Driver Know the Order to Deliver Packages? 397Why Is the Traveling Salesperson Problem So Hard? 398Solving the Traveling Salesperson Problem 399The Traveling Salesperson Problem in the Real World 400Excel Calculations 401Chapter 54 Can Data Win a Presidential Election? 405Democratic Presidential Analytics 405The GOP Strikes Back 409Cambridge Analytica and the 2016 Election 411Excel Calculations 412Chapter 55 Can Analytics Save Our Republic? 415Arrow's Impossibility Theorem 416It's Not Easy to Pick a Winner! 417Ranked-Choice Voting 419Approval Voting 420Quadratic Voting 420Excel Calculations 421Chapter 56 Why Do I Pay Too Much on eBay? 423How Many Pennies in the Jar? 423The Importance of Asymmetric Information 424The Winner's Curse and Offshore Oil Leases 424Sports Free Agents and the Winner's Curse 425Can You Avoid the Winner's Curse? 425Excel Calculations 427Chapter 57 Can Analytics Recognize, Predict, or Write a Hit Song? 429How Does Shazam Know What Song You are Listening To? 430How Did Hit Song Science Know Norah Jones's Album Would Be a Smash? 431Can Artifi cial Intelligence Write a Good Song? 433Chapter 58 Can an Algorithm Improve Parole Decisions? 437An Example of Risk Scores 438ProPublica Criticizes Risk Scores 441Skeem and Lowenkamp and PCRA 443Machine Learning and Parole Decisions 444Chapter 59 How Do Baseball Teams Decide Where to Shift Fielders? 449The Debut of the Shift 449The Return of the Shift 450Empirical Evidence on the Shift 452Why Not Just Beat the Shift? 452Excel Calculations 453Chapter 60 Did Analytics Help the Mavericks Win the 2011 NBA Title? 457How Can You Evaluate a Basketball Player? 457From Player Ratings to Lineup Ratings 459Chapter 61 Who Gets the House in the Hamptons? 463The Basic Idea 464What Asset Division Is Best? 465Excel Calculations 466Index 469

O autorze

Wayne L. Winston is Professor Emeritus of Decision Sciences at the Kelley School of Business at Indiana University. He currently teaches Sports Analytics at IU, and has taught analytics to organizations including Microsoft, eBay, Cisco, Deloitte, the U.S. military, Eli Lilly, JP Morgan, and more. A two-time Jeopardy! Champion, he has consulted on analytics for two top NBA teams.