Mostly Harmless Econometrics

From Joshua Angrist, winner of the Nobel Prize in Economics, and Jörn-Steffen Pischke, an irreverent guide to the essentials of econometricsThe core methods in today's econometric toolkit are linear regression for statistical control, instrumental variables methods for the analysis of natural experiments, and differences-in-differences methods that exploit policy changes. In the modern experimentalist paradigm, these techniques address clear causal questions such as: Do smaller classes increase learning? Should wife batterers be arrested? How much does education raise wages? Mostly Harmless Econometrics shows how the basic tools of applied econometrics allow the data to speak.In addition to econometric essentials, Mostly Harmless Econometrics covers important new extensions-regression-discontinuity designs and quantile regression-as well as how to get standard errors right. Joshua Angrist and Jörn-Steffen Pischke explain why fancier econometric techniques are typically unnecessary and even dangerous. The applied econometric methods emphasized in this book are easy to use and relevant for many areas of contemporary social science.

List of Figures viiList of Tables ixPreface xiAcknowledgments xvOrganization of This Book xviiPART I: PRELIMINARIES 1Chapter 1: Questions about Questions 3Chapter 2: The Experimental Ideal 112.1 The Selection Problem 122.2 Random Assignment Solves the Selection Problem 152.3 Regression Analysis of Experiments 22PART II: THE CORE 25Chapter 3: Making Regression Make Sense 273.1 Regression Fundamentals 283.2 Regression and Causality 513.3 Heterogeneity and Nonlinearity 683.4 Regression Details 913.5 Appendix: Derivation of the Average Derivative Weighting Function 110Chapter 4: Instrumental Variables in Action: Sometimes You Get What You Need 1134.1 IV and Causality 1154.2 Asymptotic 2SLS Inference 1384.3 Two-Sample IV and Split-Sample IV 1474.4 IV with Heterogeneous Potential Outcomes 1504.5 Generalizing LATE 1734.6 IV Details 1884.7 Appendix 216Chapter 5: Parallel Worlds: Fixed Effects, Differences-in-Differences, and Panel Data 2215.1 Individual Fixed Effects 2215.2 Differences-in-Differences 2275.3 Fixed Effects versus Lagged Dependent Variables 2435.4 Appendix: More on Fixed Effects and Lagged Dependent Variables 246PART III: EXTENSIONS 249Chapter 6: Getting a Little Jumpy: Regression Discontinuity Designs 2516.1 Sharp RD 2516.2 Fuzzy RD Is IV 259Chapter 7: Quantile Regression 2697.1 The Quantile Regression Model 2707.2 IV Estimation of Quantile Treatment Effects 283Chapter 8: Nonstandard Standard Error Issues 2938.1 The Bias of Robust Standard Error Estimates 2948.2 Clustering and Serial Correlation in Panels 3088.3 Appendix: Derivation of the Simple Moulton Factor 323Last Words 327Acronyms and Abbreviations 329Empirical Studies Index 335References 339Index 361