Micro-Econometrics for Policy, Program, and Treatment Effects (Advanced Texts in Econometrics)

Contents......Page 10
1 Tour of the book......Page 16
2.1.1 Potential outcomes and intervention......Page 22
2.1.2 Causality and association......Page 24
2.1.3 Partial equilibrium analysis and remarks......Page 25
2.2.1 Various effects......Page 26
2.2.2 Three no-effect concepts......Page 28
2.2.3 Further remarks......Page 29
2.3.1 Group-mean difference and mean effect......Page 31
2.3.2 Consequences of randomization......Page 33
2.3.3 Checking out covariate balance......Page 34
2.4.1 Overt and hidden biases......Page 36
2.4.2 Selection on observables and unobservables......Page 37
2.4.3 Linear models and biases......Page 40
2.5.1 Group-mean difference and LSE......Page 41
2.5.2 A job-training example......Page 43
2.5.3 Linking counter-factuals to linear models......Page 45
2.6 Structural form equations and treatment effect......Page 47
2.7.1 Independence and conditional independence......Page 50
2.7.2 Symmetric and asymmetric mean-independence......Page 51
2.7.3 Joint and marginal independence......Page 52
2.8.1 Illustration of biases......Page 53
2.8.2 Source of overt bias......Page 55
2.8.3 Simpson’s Paradox......Page 56
3.1 Variables to control for......Page 58
3.1.1 Must cases......Page 59
3.1.2 No-no cases......Page 60
3.1.3 Yes/no cases......Page 61
3.1.4 Option case......Page 62
3.1.5 Proxy cases......Page 63
3.2.1 Comparison group bias......Page 64
3.2.2 Dimension and support problems in conditioning......Page 66
3.2.3 Parametric models to avoid dimension and support problems......Page 68
3.2.4 Two-stage method for a semi-linear model*......Page 69
3.3.1 Parametric regression discontinuity......Page 71
3.3.2 Sharp nonparametric regression discontinuity......Page 73
3.3.3 Fuzzy nonparametric regression discontinuity......Page 76
3.3.4 Before-after (BA)......Page 79
3.4 Treatment effect estimator with weighting*......Page 80
3.4.1 Effect on the untreated......Page 82
3.4.2 Effects on the treated and on the population......Page 83
3.4.3 Effciency bounds and effcient estimators......Page 84
3.4.4 An empirical example......Page 86
3.5.1 Discrete covariates......Page 87
3.5.2 Continuous or mixed (continuous or discrete) covariates......Page 89
3.5.3 An empirical example......Page 91
4 Matching......Page 94
4.1.1 Effects on the treated......Page 95
4.1.2 Effects on the population......Page 97
4.1.3 Estimating asymptotic variance......Page 99
4.2.1 Decisions to make in matching......Page 100
4.2.2 Evaluating matching success......Page 103
4.2.3 Empirical examples......Page 105
4.3 Propensity score matching......Page 107
4.3.2 Removing overt bias with propensity-score......Page 108
4.3.3 Empirical examples......Page 110
4.4 Matching for hidden bias......Page 112
4.5.1 Mixture of before-after and matching......Page 114
4.5.2 DD for post-treatment treated in no-mover panels......Page 115
4.5.3 DD with repeated cross-sections or panels with movers......Page 118
4.5.4 Linear models for DD......Page 120
4.5.5 Estimation of DD......Page 123
4.6 Triple differences (TD)*......Page 126
4.6.1 TD for qualified post-treatment treated......Page 127
4.6.2 Linear models for TD......Page 128
4.6.3 An empirical example......Page 130
5.1 Conditions for zero hidden bias......Page 132
5.2.1 Partial treatment......Page 134
5.2.2 Reverse treatment......Page 137
5.3 Multiple responses......Page 138
5.4 Multiple control groups......Page 140
5.5.1 Potential treatments......Page 144
5.5.2 Sources for instruments......Page 146
5.5.3 Relation to regression discontinuity design......Page 149
5.6.1 Wald estimator under constant effects......Page 151
5.6.2 IVE for heterogenous effects......Page 153
5.6.3 Wald estimator as effect on compliers......Page 154
5.6.4 Weighting estimators for complier effects*......Page 157
6.1 Sensitivity analysis......Page 162
6.1.1 Unobserved confounder affecting treatment......Page 163
6.1.2 Unobserved confounder affecting treatment and response......Page 167
6.1.3 Average of ratios of biased to true effects......Page 172
6.2 Selection correction methods......Page 175
6.3 Nonparametric bounding approaches......Page 178
6.4 Controlling for post-treatment variables to avoid confounder......Page 182
7.1 Multiple treatments......Page 186
7.1.1 Parameters of interest......Page 187
7.1.2 Balancing score and propensity score matching......Page 189
7.2 Treatment duration effects with time-varying covariates......Page 192
7.3.1 Motivation with two-period linear models......Page 196
7.3.2 G algorithm under no unobserved confounder......Page 201
7.3.3 G algorithm for three or more periods......Page 203
A.1 Kernel nonparametric regression......Page 206
A.2.1 Comparison to a probabilistic causality......Page 211
A.2.2 Learning about joint distribution from marginals......Page 213
A.3.1 Derivation for a semi-linear model......Page 216
A.3.2 Derivation for weighting estimators......Page 217
A.4.1 Non-sequential matching with network flow algorithm......Page 219
A.4.2 Greedy non-sequential multiple matching......Page 221
A.4.3 Nonparametric matching and support discrepancy......Page 224
A.5.1 Some remarks on LATE......Page 229
A.5.2 Outcome distributions for compliers......Page 231
A.5.3 Median treatment effect......Page 234
A.6.1 Controlling for affected covariates in a linear model......Page 236
A.6.2 Controlling for affected mean-surrogates......Page 239
A.7.1 Regression models for discrete cardinal treatments......Page 241
A.7.2 Complete pairing for censored responses......Page 243
References......Page 248
E......Page 260
M......Page 261
S......Page 262
Z......Page 263