Samprit Chatterjee - Handbook of Regression Analysis With Applications in R

1 Cover

2 Preface to the Second Edition

3 Preface to the First EditionHow to Use This Book

4 PART ONE: The Multiple Linear Regression Model CHAPTER ONE: Multiple Linear Regression 1.1 Introduction 1.2 Concepts and Background Material 1.3 Methodology 1.4 Example — Estimating Home Prices 1.5 Summary CHAPTER TWO: Model Building 2.1 Introduction 2.2 Concepts and Background Material 2.3 Methodology 2.4 Indicator Variables and Modeling Interactions 2.5 Summary

5 PART TWO: Addressing Violations of Assumptions CHAPTER THREE: Diagnostics for Unusual Observations 3.1 Introduction 3.2 Concepts and Background Material 3.3 Methodology 3.4 Example — Estimating Home Prices (continued) 3.5 Summary CHAPTER FOUR: Transformations and Linearizable Models 4.1 Introduction 4.2 Concepts and Background Material: The Log‐Log Model 4.3 Concepts and Background Material: Semilog Models 4.4 Example — Predicting Movie Grosses After One Week 4.5 Summary CHAPTER FIVE: Time Series Data and Autocorrelation 5.1 Introduction 5.2 Concepts and Background Material 5.3 Methodology: Identifying Autocorrelation 5.4 Methodology: Addressing Autocorrelation 5.5 Summary

6 PART THREE: Categorical Predictors CHAPTER SIX: Analysis of Variance 6.1 Introduction 6.2 Concepts and Background Material 6.3 Methodology 6.4 Example — DVD Sales of Movies 6.5 Higher‐Way ANOVA 6.6 Summary CHAPTER SEVEN: Analysis of Covariance 7.1 Introduction 7.2 Methodology 7.3 Example — International Grosses of Movies 7.4 Summary

7 PART FOUR: Non‐Gaussian Regression Models CHAPTER EIGHT: Logistic Regression 8.1 Introduction 8.2 Concepts and Background Material 8.3 Methodology 8.4 Example — Smoking and Mortality 8.5 Example — Modeling Bankruptcy 8.6 Summary CHAPTER NINE: Multinomial Regression 9.1 Introduction 9.2 Concepts and Background Material 9.3 Methodology 9.4 Example — City Bond Ratings 9.5 Summary CHAPTER TEN: Count Regression 10.1 Introduction 10.2 Concepts and Background Material 10.3 Methodology 10.4 Overdispersion and Negative Binomial Regression 10.5 Example — Unprovoked Shark Attacks in Florida 10.6 Other Count Regression Models 10.7 Poisson Regression and Weighted Least Squares 10.8 Summary CHAPTER ELEVEN: Models for Time‐to‐Event (Survival) Data 11.1 Introduction 11.2 Concepts and Background Material 11.3 Methodology 11.4 Example — The Survival of Broadway Shows (continued) 11.5 Left‐Truncated/Right‐Censored Data and Time‐Varying Covariates 11.6 Summary

8 PART FIVE: Other Regression Models CHAPTER TWELVE: Nonlinear Regression 12.1 Introduction 12.2 Concepts and Background Material 12.3 Methodology 12.4 Example — Michaelis‐Menten Enzyme Kinetics 12.5 Summary CHAPTER THIRTEEN: Models for Longitudinal and Nested Data 13.1 Introduction 13.2 Concepts and Background Material 13.3 Methodology 13.4 Example — Tumor Growth in a Cancer Study 13.5 Example — Unprovoked Shark Attacks in the United States 13.6 Summary CHAPTER FOURTEEN: Regularization Methods and Sparse Models 14.1 Introduction 14.2 Concepts and Background Material 14.3 Methodology 14.4 Example — Human Development Index 14.5 Summary

9 PART SIX: Nonparametric and Semiparametric Models CHAPTER FIFTEEN: Smoothing and Additive Models 15.1 Introduction 15.2 Concepts and Background Material 15.3 Methodology 15.4 Example — Prices of German Used Automobiles 15.5 Local and Penalized Likelihood Regression 15.6 Using Smoothing to Identify Interactions 15.7 Summary CHAPTER SIXTEEN: Tree‐Based Models 16.1 Introduction 16.2 Concepts and Background Material 16.3 Methodology 16.4 Examples 16.5 Trees for Other Types of Data 16.6 Summary

10 Bibliography

11 Index

12 End User License Agreement

1 Chapter 2 Table 2.1: Variance inflation caused by correlation of predictors in a two‐pred...

2 Chapter 4Table 4.1: Total domestic grosses (in millions of dollars) for 2010 movies with

3 Chapter 6Table 6.1: Indicator variable codings for a one‐way ANOVA fit omitting the vari...Table 6.2: Effect codings for a one‐way ANOVA fit omitting the variable for lev...Table 6.3: Indicator variable codings for a one‐way ANOVA fit with omitting th...Table 6.4: Effect codings for a one‐way ANOVA fit with omitting the variable f...

4 Chapter 8Table 8.1: Details of models fit to the Whickham smoking study data.Table 8.2: Details of models fit to the bankruptcy data with 360Networks omitte...

5 Chapter 11Table 11.1: Log‐rank test statistics for Broadway show post‐Tony Awards perform...Table 11.2: Time‐varying covariates survival data.

6 Chapter 13Table 13.1: Estimated expected logged tumor size and growth rate of logged tumo...

7 Chapter 14Table 14.1: Coefficient estimates of selected models using different methods (l...

1 Chapter 1 FIGURE 1.1: The simple linear regression model. The solid line corresponds to ... FIGURE 1.2: Least squares estimation for the simple linear regression model, u... FIGURE 1.3: Least squares estimation for the multiple linear regression model ... FIGURE 1.4: Scatter plots of sale price versus each predictor for the home pri... FIGURE 1.5: Residual plots for the home price data. (a) Plot of residuals vers... FIGURE 1.6: Scatter plots of residuals versus each predictor for the home pric...

2 Chapter 2 FIGURE 2.1: Least squares estimation under collinearity. The only change in th... FIGURE 2.2: Residual plots for the home price data using the best three‐predic... FIGURE 2.3: Plot of observed versus predicted house sale price values of valid... FIGURE 2.4: Plots for the 2004 election data. (a) Plot of percentage change in... FIGURE 2.5: Regression lines for election data separated by whether the county... FIGURE 2.6: Plots for the 2004 election data. (a) Plot of percentage change in... FIGURE 2.7: Residual plots for the 2004 election data.

3 Chapter 3 FIGURE 3.1: Scatter plot of data with an unusual observation (A), with the fit... FIGURE 3.2: Scatter plots of data with an unusual observation, with the fitted... FIGURE 3.3: Index plots of diagnostics for the three‐predictor regression fit ...FIGURE 3.4: Plot of Cook's distance versus diagonal element of the hat matrix ...FIGURE 3.5: Index plots of diagnostics for the two‐predictor regression fit fo...

4 Chapter 4FIGURE 4.1: Scatter plots of total domestic gross versus opening weekend gross...FIGURE 4.2: Scatter plots of logged total domestic gross versus logged opening...FIGURE 4.3: Index plots of diagnostics for the regression fit on the three‐pre...FIGURE 4.4: Index plots of diagnostics for the three‐predictor regression fit ...FIGURE 4.5: Residual plots for the three‐predictor regression fit after removi...

5 Chapter 5FIGURE 5.1: Inefficiency of OLS estimator compared to minimum variance estimat...FIGURE 5.2: Percentage bias of the estimated variance of the OLS estimator of FIGURE 5.3: Time series plot of quarterly U.S. e‐commerce retail sales.FIGURE 5.4: Scatter plot of e‐commerce retail sales versus total retail sales.FIGURE 5.5: Residual plots for the e‐commerce regression fit based on total re...FIGURE 5.6: Side‐by‐side boxplots of the standardized residuals for the e‐comm...FIGURE 5.7: Residual plots for the e‐commerce regression fit based on total re...FIGURE 5.8: Residual plots for the e‐commerce regression fit based on total re...FIGURE 5.9: Scatter plots of daily S&P 500 (large‐cap) index values from 2006 ...FIGURE 5.10: Scatter plots of logged daily S&P 500 (large‐cap) index values ve...FIGURE 5.11: Residual plots for logged large‐cap index regression fit based on...FIGURE 5.12: Residual plots for the logged large‐cap index regression fit base...FIGURE 5.13: Residual plots for the large‐cap log return regression fit based ...FIGURE 5.14: Scatter plot of the time interval to next eruption versus the dur...FIGURE 5.15: Residual plots for the Old Faithful Geyser time interval until th...FIGURE 5.16: Residual plots for the Old Faithful Geyser time interval until th...