본문 바로가기 주메뉴 바로가기
국회도서관 홈으로 정보검색 소장정보 검색

목차보기

Title page

Contents

ABSTRACT 3

1. INTRODUCTION 4

2. NONPARAMETRIC DIFFERENCE-IN-DIFFERENCES 6

2.1. Difference-in-differences with covariates 6

2.2. Nonparametric conditional expectations 8

3. COVARIATES AND SCALE 9

3.1. Data-driven evaluation of potential scales 10

3.2. Data-driven evaluation of confounder sets 11

4. TREATMENT EFFECT ESTIMATORS 12

4.1. Conditional treatment effect on the treated 12

4.2. Unconditional treatment effect on the treated 13

5. TESTING 14

5.1. Significance of treatment effects 14

5.2. Composite significance testing in model-free DiD 14

5.3. Asymptotic behavior 15

5.4. Feasible bootstrap tests 16

6. APPLICATION: HUMAN CAPITAL RESPONSES TO DACA 17

6.1. Data 17

6.2. Empirical results 18

7. CONCLUSIONS AND DIRECTIONS FOR FURTHER EXTENSIONS 21

REFERENCES 22

A. APPENDIX 23

SUPPLEMENT FOR ORIGINAL ARTICLE 27

Tables

Table 1. effect of DACA on school attendance 19

Table 2. effect of DACA on high school completion and college enrollment 20

Supplement Tables

Table 1. Fraction correctly choosing S₁,₂ versus alternative sets of covariates: AIC penalty factor included, average sample size (to the nearest integer)... 32

Table 2. Size and power of our bias stability ('parallel path') condition test (T0): The probability of rejection at each significance level (1, 5 and 10%)... 34

Table 3. Performance of nonparametric TTb estimator: Average bias and MSE over the simulations and average variance (calculated via B = 999... 34