Identification Problems in the Social Sciences by Charles F. Manski -- Read -- Imperial Library of Trantor

Index

Cover Title Copyright Dedication Contents Preface Introduction

Identification Tolerating Ambiguity

1 Extrapolation

1.1 Predicting Criminality 1.2 Probabilistic Prediction 1.3 Inferring Conditional Distributions from Random-Sample Data 1.4 Prior Distributional Information 1.5 Predicting High School Graduation

2 The Selection Problem

2.1 The Nature of the Problem 2.2 Identification from Censored Samples Alone 2.3 Bounding the Probability of Exiting Homelessness 2.4 Prior Distributional Information 2.5 Identification of Treatment Effects 2.6 Information Linking Outcomes across Treatments 2.7 Predicting High School Graduation If All Families Were Intact

3 The Mixing Problem in Program Evaluation

3.1 The Experimental Evaluation of Social Programs 3.2 Variation in Treatment 3.3 The Perry Preschool Project 3.4 Identification of Mixtures Using Only Knowledge of the Marginals 3.5 Restrictions on the Outcome Distribution 3.6 Restrictions on the Treatment Policy 3.7 Identifying Combinations of Assumptions

4 Response-Based Sampling

4.1 The Odds Ratio and Public Health 4.2 Bounds on Relative and Attributable Risk 4.3 Information on Marginal Distributions 4.4 Sampling from One Response Stratum 4.5 General Binary Stratifications

5 Predicting Individual Behavior

5.1 Revealed Preference Analysis 5.2 How Do Youth Infer the Returns to Schooling? 5.3 Analysis of Intentions Data

6 Simultaneity

6.1 “The” Identification Problem in Econometrics 6.2 The Linear Market Model 6.3 Equilibrium in Games 6.4 Simultaneity with Downward-Sloping Demand

7 The Reflection Problem

7.1 Endogenous, Contextual, and Correlated Effects 7.2 A Linear Model 7.3 A Pure Endogenous Effects Model 7.4 Inferring the Composition of Reference Groups 7.5 Dynamic Analysis

Notes References Name Index Subject Index