Identification for Prediction and Decision / Edition 1

by Charles F. Manski

Overview

This book is a full-scale exposition of Charles Manski's new methodology for analyzing empirical questions in the social sciences. He recommends that researchers first ask what can be learned from data alone, and then ask what can be learned when data are combined with credible weak assumptions. Inferences predicated on weak assumptions, he argues, can achieve wide consensus, while ones that require strong assumptions almost inevitably are subject to sharp disagreements.

Building on the foundation laid in the author's Identification Problems in the Social Sciences (Harvard, 1995), the book's fifteen chapters are organized in three parts. Part I studies prediction with missing or otherwise incomplete data. Part II concerns the analysis of treatment response, which aims to predict outcomes when alternative treatment rules are applied to a population. Part III studies prediction of choice behavior.

Each chapter juxtaposes developments of methodology with empirical or numerical illustrations. The book employs a simple notation and mathematical apparatus, using only basic elements of probability theory.

Product Details

ISBN-13:	9780674026537
Publisher:	Harvard University Press
Publication date:	01/31/2008
Edition description:	New Edition
Pages:	368
Product dimensions:	6.12(w) x 9.25(h) x 1.00(d)

About the Author

Charles F. Manski is Board of Trustees Professor of Economics at Northwestern University.

Table of Contents

Preface     xiii
Introduction     1
The Reflection Problem     1
The Law of Decreasing Credibility     2
Identification and Statistical Inference     3
Prediction and Decisions     6
Coping with Ambiguity     6
Organization of the Book     8
The Developing Literature on Partial Identification     11
Prediction with Incomplete Data
Conditional Prediction     17
Predicting Criminality     17
Probabilistic Prediction     18
Estimation of Best Predictors from Random Samples     22
Extrapolation     25
Predicting High School Graduation     28
Best Predictors under Square and Absolute Loss     30
Nonparametric Regression Analysis     32
Word Problems     34
Missing Outcomes     36
Anatomy of the Problem     37
Bounding the Probability of Exiting Homelessness     40
Means of Functions of the Outcome     42
Parameters That Respect Stochastic Dominance     44
Distributional Assumptions     45
Wage Regressions and the Reservation-Wage Model of Labor Supply     48
Statistical Inference     51
Interval Measurement of Outcomes     54
Jointly Missing Outcomes and Covariates     56
Convergence of Sets to Sets     60
Instrumental Variables     62
Distributional Assumptions and Credible Inference     62
Missingness at Random     64
Statistical Independence     66
Equality of Means     69
Inequality of Means     71
Imputations and Nonresponse Weights     73
Conditioning on the Propensity Score     75
Word Problems     76
Parametric Prediction     83
The Normal-Linear Model of Market and Reservation Wages     83
Selection Models     87
Parametric Models for Best Predictors     89
Minimum-Distance Estimation of Partially Identified Models     91
Decomposition of Mixtures     94
The Inferential Problem and Some Manifestations     94
Binary Mixing Covariates     98
Contamination through Imputation     102
Instrumental Variables     105
Sharp Bounds on Parameters That Respect Stochastic Dominance     107
Response-Based Sampling     109
The Odds Ratio and Public Health     110
Bounds on Relative and Attributable Risk     114
Information on Marginal Distributions     118
Sampling from One Response Stratum     119
General Binary Stratifications     122
Analysis of Treatment Response
The Selection Problem     127
Anatomy of the Problem     128
Sentencing and Recidivism     134
Randomized Experiments     136
Compliance with Treatment Assignment     140
Treatment by Choice     148
Treatment at Random in Nonexperimental Settings     151
Homogeneous Linear Response     153
Perspectives on Treatment Comparison     157
Word Problems     160
Linear Simultaneous Equations     167
Simultaneity in Competitive Markets     167
The Linear Market Model     170
Equilibrium in Games     174
The Reflection Problem     177
Monotone Treatment Response     183
Shape Restrictions     183
Bounds on Parameters That Respect Stochastic Dominance     186
Bounds on Treatment Effects     189
Monotone Response and Selection     191
Bounding the Returns to Schooling     193
The Mixing Problem      198
Extrapolation from Experiments to Rules with Treatment Variation     198
Extrapolation from the Perry Preschool Experiment     200
Identification of Event Probabilities with the Experimental Evidence Alone     204
Treatment Response Assumptions     206
Treatment Rule Assumptions     207
Combining Assumptions     210
Planning under Ambiguity     211
Studying Treatment Response to Inform Treatment Choice     211
Criteria for Choice under Ambiguity     214
Treatment Using Data from an Experiment with Partial Compliance     218
An Additive Planning Problem     222
Planning with Partial Knowledge of Treatment Response     226
Planning and the Selection Problem     229
The Ethics of Fractional Treatment Rules     233
Decentralized Treatment Choice     235
Minimax-Regret Rules for Two Treatments Are Fractional     237
Reporting Observable Variation in Treatment Response     239
Word Problems     241
Planning with Sample Data     243
Statistical Induction     243
Wald's Development of Statistical Decision Theory     245
Using a Randomized Experiment to Evaluate an Innovation     250
Predicting Choice Behavior
Revealed Preference Analysis     259
Revealing the Preferences of an Individual     260
Random Utility Models of Population Choice Behavior     263
College Choice in America     270
Random Expected-Utility Models     274
Prediction Assuming Strict Preferences     278
Axiomatic Decision Theory     282
Measuring Expectations     284
Elicitation of Expectations from Survey Respondents     285
Illustrative Findings     290
Using Expectations Data to Predict Choice Behavior     295
Measuring Ambiguity     298
The Predictive Power of Intentions Data: A Best-Case Analysis     300
Measuring Expectations of Facts     305
Studying Human Decision Processes     308
As-If Rationality and Bounded Rationality     309
Choice Experiments     312
Prospects for a Neuroscientific Synthesis     317
References     321
Author Index     339
Subject Index     343

What People are Saying About This

Charles Manski is a highly original and influential voice in econometrics. His work on partial identification and nonparametric bounds now holds a central position in many areas of theoretical and applied research. This comprehensive yet accessible text brings together the author's research on incomplete data, on treatment response and on choice behavior. It is an important contribution to our knowledge and will stand as a key reference for students and researchers for years to come.

Editorial Reviews

Charles Manski is a highly original and influential voice in econometrics. His work on partial identification and nonparametric bounds now holds a central position in many areas of theoretical and applied research. This comprehensive yet accessible text brings together the author's research on incomplete data, on treatment response and on choice behavior. It is an important contribution to our knowledge and will stand as a key reference for students and researchers for years to come.

Richard Blundell

More than anyone else, Charles Manski has changed the way we think about identification. This book contains the most comprehensive discussion of his work in this area. It is a must-read for everybody interested in identification, and there isn't an empirical economist or econometrician who can afford not to be.

Guido Imbens