Muutke küpsiste eelistusi

Matching, Regression Discontinuity, Difference in Differences, and Beyond [Pehme köide]

5.00/5 (1 hinnangut Goodreads-ist)
(Professor of Economics, Korea University, Seoul)
  • Formaat: Paperback / softback, 280 pages, kõrgus x laius x paksus: 231x155x13 mm, kaal: 358 g
  • Ilmumisaeg: 23-Jun-2016
  • Kirjastus: Oxford University Press Inc
  • ISBN-10: 0190258748
  • ISBN-13: 9780190258740
Teised raamatud teemal:
Matching, Regression Discontinuity, Difference in Differences, and BeyondSuurem pilt
  • Formaat: Paperback / softback, 280 pages, kõrgus x laius x paksus: 231x155x13 mm, kaal: 358 g
  • Ilmumisaeg: 23-Jun-2016
  • Kirjastus: Oxford University Press Inc
  • ISBN-10: 0190258748
  • ISBN-13: 9780190258740
Teised raamatud teemal:
Püsilink: https://www.kriso.ee/db/9780190258740.html
Märksõnad:
This book reviews the three most popular methods (and their extensions) in applied economics and other social sciences: matching, regression discontinuity, and difference in differences. The book introduces the underlying econometric/statistical ideas, shows what is identified and how the identified parameters are estimated, and then illustrates how they are applied with real empirical examples. The book emphasizes how to implement the three methods with data: many data and programs are provided in the online appendix. All readers---theoretical econometricians/statisticians, applied economists/social-scientists and researchers/students---will find something useful in the book from different perspectives.

Arvustused

The book under review gives an up-to-date and concise treatment of the three most recent and widely applied empirical methods to deal with confoundedness in microeconometrics, i.e., matching, regression discontinuity and difference in differences. This book gives a timely update on the author's previous book Microeconometrics for Policy, Program, and Treatment Effects with a particular concentration on these three most in fluential methodologies. This book serves as an excellent reference for both graduate students and researchers working in the field of microeconometrics. * Zhongwen Liang, Mathematical Reviews Clippings *

Preface xv
1 Basics of Treatment Effect Analysis
1(27)
1.1 Counterfactual, Intervention, and Causal Relation
1(4)
1.1.1 Potential Outcomes and Intervention
1(2)
1.1.2 Causality and Association
3(1)
1.1.3 Partial Equilibrium Analysis and Remarks
4(1)
1.2 Various Treatment Effects and No Effects
5(3)
1.2.1 Various Effects
5(1)
1.2.2 Three No-Effect Concepts
6(1)
1.2.3 Remarks
7(1)
1.3 Group-Mean Difference and Randomization
8(5)
1.3.1 Group-Mean Difference and Mean Effect
8(2)
1.3.2 Consequences of Randomization
10(2)
1.3.3 Checking Out Covariate Balance
12(1)
1.4 Overt Bias, Hidden Bias, and Selection Problems
13(5)
1.4.1 Overt and Hidden Biases
13(1)
1.4.2 Selection on Observables and Unobservables
14(2)
1.4.3 Linear Models and Biases
16(2)
1.5 Estimation with Group Mean Difference and LSE
18(4)
1.5.1 Group-Mean Difference and LSE
18(1)
1.5.2 Job Training Example
19(2)
1.5.3 Linking Counterfactuals to Linear Models
21(1)
1.6 Structural Form, Assignment, and Marginal Model
22(3)
1.6.1 Structural versus Reduced Forms for Response
22(1)
1.6.2 Treatment Structural Form and Assignment
23(1)
1.6.3 Marginal Structural Model
24(1)
1.7 Simpson's Paradox and False Covariate Control
25(3)
2 Matching
28(33)
2.1 Basics of Matching and Various Effects
28(7)
2.1.1 Main Idea
28(2)
2.1.2 Effect on Treated and Effect on Population
30(1)
2.1.3 Dimension and Support Problems
31(1)
2.1.4 Variables to Control
32(3)
2.2 Implementing Matching
35(11)
2.2.1 Decisions to Make in Matching
35(2)
2.2.2 Matching Estimators
37(3)
2.2.3 Asymptotic Variance Estimation
40(4)
2.2.4 Labor Union Effect on Wage
44(2)
2.3 Propensity Score Matching (PSM)
46(8)
2.3.1 Propensity Score as a Balancing Score
46(1)
2.3.2 Removing Overt Bias with Propensity Score
47(1)
2.3.3 Implementing PSM and Bootstrap
48(2)
2.3.4 PSM Empirical Examples
50(2)
2.3.5 Propensity Score Specification Issues*
52(2)
2.4 Further Remarks
54(7)
2.4.1 Covariate Balance Check
54(2)
2.4.2 Matching for Hidden Bias
56(2)
2.4.3 Prognostic Score and More*
58(3)
3 Nonmatching and Sample Selection
61(36)
3.1 Weighting
61(8)
3.1.1 Weighting Estimator for Effect on Population
61(2)
3.1.2 Other Weighting Estimators and Remarks
63(2)
3.1.3 Asymptotic Distribution of Weighting Estimators*
65(1)
3.1.4 Job Training Effect on Unemployment
66(1)
3.1.5 Doubly Robust Estimator*
67(1)
3.1.6 Weighting for Missing Data*
68(1)
3.2 Regression Imputation
69(7)
3.2.1 Linear Regression Imputation
70(1)
3.2.2 Regression Imputation with Propensity Score
71(1)
3.2.3 Regression Imputation for Multiple Treatment
72(1)
3.2.4 Regression Imputation for Continuous Treatment*
73(1)
3.2.5 Military Service Effect on Wage
74(2)
3.3 Complete Pairing with Double Sum
76(8)
3.3.1 Discrete Covariates
77(2)
3.3.2 Continuous Covariates
79(1)
3.3.3 Nonparametric Distributional Effect Tests*
80(4)
3.4 Treatment Effects under Sample Selection
84(6)
3.4.1 Difficulties with Sample Selection Models
85(1)
3.4.2 Participation, Invisible, and Visible Effects
86(1)
3.4.3 Identification of Three Effects with Mean Differences
87(1)
3.4.4 Religiosity Effect on Affairs
88(2)
3.5 Effect Decomposition in Sample Selection Models*
90(7)
3.5.1 Motivation for Decomposition
90(1)
3.5.2 Decomposition with Linear Selection Model
91(1)
3.5.3 Four Special Models
92(2)
3.5.4 Race Effect on Wage
94(3)
4 Regression Discontinuity
97(34)
4.1 Introducing RD with Before-After
97(3)
4.1.1 BA Examples
97(1)
4.1.2 BA Identification Assumption
98(1)
4.1.3 From BA to RD
99(1)
4.2 RD Identification and Features
100(9)
4.2.1 Sharp RD (SRD) and Fuzzy RD (FRD)
101(1)
4.2.2 Identification at Cutoff
102(2)
4.2.3 RD Main Features
104(2)
4.2.4 Class Size Effect on Test Score
106(3)
4.3 RD Estimators
109(7)
4.3.1 LSE for Level Equation
109(1)
4.3.2 IVE for Right-Left Differenced Equation
110(2)
4.3.3 Bandwidth Choice and Remarks
112(1)
4.3.4 High School Completion Effect on Fertility
113(3)
4.4 Specification Tests
116(3)
4.4.1 Breaks in Conditional Means
116(1)
4.4.2 Continuity in Score Density
117(2)
4.5 RD Topics*
119(12)
4.5.1 Spatial Breaks
119(1)
4.5.2 RD for Limited Dependent Variables
120(1)
4.5.3 Measurement Error in Score
121(2)
4.5.4 Regression Kink (RK) and Generalization
123(3)
4.5.5 SRD with Multiple Scores
126(3)
4.5.6 Quantile RD
129(2)
5 Difference in Differences
131(34)
5.1 DD Basics
131(5)
5.1.1 Examples for DD
132(1)
5.1.2 Time-Constant and Time-Varying Qualifications
133(2)
5.1.3 Data Requirement and Notation
135(1)
5.2 DD with Repeated Cross-Sections
136(14)
5.2.1 Identification
136(4)
5.2.2 Identification with Parametric Models
140(2)
5.2.3 Schooling Effect on Fertility: 'Fuzzy DD
142(2)
5.2.4 Linear Model Estimation for Two Periods or More
144(3)
5.2.5 Earned Income Tax Credit Effect on Work
147(1)
5.2.6 Time-Varying Qualification*
148(2)
5.3 DD with Panel Data
150(8)
5.3.1 Identification
150(2)
5.3.2 Identification and Estimation with Parametric Models
152(5)
5.3.3 Daylight Saving Time Effect on Energy
157(1)
5.4 Panel Stayer DD for Time-Varying Qualification
158(7)
5.4.1 Motivation
158(1)
5.4.2 Effect on In- Stayers Identified by Stayer DD
159(1)
5.4.3 Identification and Estimation with Panel Linear Models
160(2)
5.4.4 Pension Effect on Health Expenditure
162(3)
6 Triple Difference and Beyond
165(44)
6.1 TD Basics and More
165(1)
6.2 TD with Repeated Cross-Sections
166(8)
6.2.1 Identification
166(3)
6.2.2 Identification and Estimation with Linear Models
169(3)
6.2.3 Mandated Benefit Effect on Wage
172(2)
6.3 TD with Panel Data
174(4)
6.3.1 Identification
174(1)
6.3.2 Estimation with Panel Linear Model
175(2)
6.3.3 Tax-Inclusive Price Effect on Demand
177(1)
6.4 GDD and Beyond
178(9)
6.4.1 Motivation for GDD and Beyond
179(1)
6.4.2 Identification for GDD and QD
180(1)
6.4.3 Identified Effects When Panel Linear Model Holds
181(1)
6.4.4 LSE for DD and GDD and Testing for DD Condition
182(2)
6.4.5 Sulfa Drug Effect on Mortality: Is DD Trustworthy?
184(3)
6.5 Clustering Problems and Inference for DD and TD
187(22)
6.5.1 Single Clustering
188(6)
6.5.2 Clustering in Panel Data
194(5)
6.5.3 DD and TD with Cluster- Specific Treatment
199(3)
6.5.4 Details on Cluster Variance Estimator*
202(7)
A Appendix
209(32)
A.1 Kernel Density and Regression Estimators
209(4)
A.1.1 Histogram-Type Density Estimator
209(1)
A.1.2 Kernel Density Estimator
210(1)
A.1.3 Kernel Regression Estimator
211(2)
A.1.4 Local Linear Regression
213(1)
A.2 Bootstrap
213(7)
A.2.1 Review on Usual Asymptotic Inference
215(1)
A.2.2 Bootstrap to Find Quantiles
216(2)
A.2.3 Percentile-t and Percentile Methods
218(1)
A.2.4 Nonparametric, Parametric, and Wild Bootstraps
219(1)
A.3 Confounder Detection, IVE, and Selection Correction
220(12)
A.3.1 Coherence Checks
220(5)
A.3.2 WE and Complier Effect
225(5)
A.3.3 Selection Correction Approach
230(2)
A.4 Supplements for DD
Chapter
232(9)
A.4.1 Nonparametric Estimators for Repeated Cross-Section DD
233(1)
A.4.2 Nonparametric Estimation for DD with Two-Wave Panel Data
233(3)
A.4.3 Panel Linear Model Estimation for DD with One-Shot Treatment
236(2)
A.4.4 Change in Changes
238(3)
References 241(12)
Index 253
Professor Myoung-jae Lee is an econometrician/statistician in Korea University. He received his Ph.D. in economics from University of Wisconsin-Madison in 1989. Since then, he published about 60 papers in economic and statistic journals as well as top-rated journals in other fields, including Econometrica, Journal of the Royal Statistical Society (Series B), Journal of Econometrics, Transportation Research (Part B), and Sociological Methods & Research. He also published four single-authored micro-econometric books from Springer, Academic Press and Oxford University Press on limited dependent variables, panel data and treatment effect analysis.