Some books are correct. Some are clear. Some are useful. Some are entertaining. Few are even two of these. This book is all four. Statistical Models: Theory and Practice is lucid, candid and insightful, a joy to read. We are fortunate that David Freedman finished this new edition before his death in late 2008. We are deeply saddened by his passing, and we greatly admire the energy and cheer he brought to this volume——and many other projects——-during his final months.
作者简介
David A.Freedman（1938-2008）是加州大学伯利分校的统计学教授、杰出的数理统计学家，其研究范围包括鞅不等式分析、Markov过程、抽样、自助法等。他是美国科学学院（American Academy of Art and Sciences）院士。在2003年，美国科学院授予他John J.Carry科学进步奖，以表彰他对统计理论和实。
目录
Table of Contents Foreword to the Revised Edition iii Preface v 1 Observational Studies and Experiments 1.1 Introduction 1 1.2 The HIP trial 4 1.3 Snow on cholera 6 1.4 Yule on the causes of poverty 9 Exercise set A 13 1.5 End notes 14
2 The Regression Line 2.1 Introduction 18 2.2 The regression line 18 2.3 Hooke's law 22 Exercise set A 23 2.4 Complexities 23 2.5 Simple vs multiple regression 26 Exercise set B 26 2.6 End notes 28
3 Matrix Algebra 3.1 Introduction 29 Exercise set A 30 3.2 Determinants and inverses 31 Exercise set B 33 3.3 Random vectors 35 Exercise set C 35 3.4 Positive definite matrices 36 Exercise set D 37 3.5 The normal distribution 38 Exercise set E 39 3.6 If you want a book on matrix algebra 40
4 Multiple Regression 4.1 Introduction 41 Exercise set A 44 4.2 Standard errors 45 Things we don't need 49 Exercise set B 49 4.3 Explained variance in multiple regression 51 Association or causation? 53 Exercise set C 53 4.4 What happens to OLS if the assumptions break down? 53 4.5 Discussion questions 53 4.6 End notes 59
5 Multiple Regression: Special Topics 5.1 Introduction 61 5.20LSisBLUE 61 Exercise set A 63 5.3 Generalized least squares 63 Exercise set B 65 5.4 Examples on GLS 65 Exercise set C 66 5.5 What happens to GLS if the assumptions break down? 68 5.6 Normal theory 68 Statistical significance 70 Exercise set D 71 5.7 The F-test 72 "The" F-test in applied work 73 Exercise set E 74 5.8 Data snooping 74 Exercise set F 76 5.9 Discussion questions 76 5.10 End notes 78
6 Path Models 6.1 Stratification 81 Exercise set A 86 6.2 Hooke's law revisited 87 Exercise set B 88 6.3 Political repression during the McCarthy era 88 Exercise set C 90 TABLE OF CONTENTS 6.4 Inferring causation by regression 91 Exercise set D 93 6.5 Response schedules for path diagrams 94 Selection vs intervention 101 Structural equations and stable parameters 101 Ambiguity in notation 102 Exercise set E 102 6.6 Dummy variables 103 Types of variables 104 6.7 Discussion questions 105 6.8 End notes 112
7 Maximum Likelihood 7.1 Introduction 115 Exercise set A 119 7.2 Probit models 121 Why not regression? 123 The latent-variable formulation 123 Exercise set B 124 Identification vs estimation 125 What if the Ui are N(/z, tr2)? 126 Exercise set C 127 7.3 Logit models 128 Exercise set D 128 7.4 The effect of Catholic schools 130 Latent variables 132 Response schedules 133 The second equation 134 Mechanics: bivariate probit 136 Why a model rather than a cross-tab? 138 Interactions 138 More on table 3 in Evans and Schwab 139 More on the second equation 139 Exercise set E 140 7.5 Discussion questions 141 7.6 End notes 150 8 The Bootstrap 8.1 Introduction 155 Exercise set A 166 8.2 Bootstrapping a model for energy demand 167 Exercise set B 173 8.3 End notes 174
9 Simultaneous Equations 9.1 Introduction 176 Exercise set A 181 9.2 Instrumental variables 181 Exercise set B 184 9.3 Estimating the butter model 184 Exercise set C 185 9.4 What are the two stages? 186 Invariance assumptions 187 9.5 A social-science example: education and fertility 187 More on Rindfuss et al 191 9.6 Covariates 192 9.7 Linear probability models 193 The assumptions 194 The questions 195 Exercise set D 196 9.8 More on IVLS 197 Some technical issues 197 Exercise set E 198 Simulations to illustrate IVLS 199 9.9 Discussion questions 200 9.10 End notes 207
10 Issues in Statistical Modeling 10.1 Introduction 209 The bootstrap 211 The role of asymptotics 211 Philosophers' stones 211 The modelers' response 212 10.2 Critical literature 212 10.3 Response schedules 217 10.4 Evaluating the models in chapters 7-9 217 10.5 Summing up 218 References 219 Answers to Exercises 235 TABLE OF CONTENTS The Computer Labs 294 Appendix: Sample MATLAB Code 310 Reprints Gibson on McCarthy 315 Evans and Schwab on Catholic Schools 343 Rindfuss et al on Education and Fertility 377 Schneider et al on Social Capital 402 Index 431
