Bounds on treatment effects from studies with imperfect compliance

被引：358

作者：

Balke, A

Pearl, J

机构：

来源：

JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION | 1997年 / 92卷 / 439期

关键词：

causal models; latent variables; linear programming; noncompliance;

D O I：

10.2307/2965583

中图分类号：

O21 [概率论与数理统计]; C8 [统计学];

学科分类号：

020208 [统计学]; 070103 [概率论与数理统计]; 0714 [统计学];

摘要：

This article establishes nonparametric formulas that can be used to bound the average treatment effect in experimental studies in which treatment assignment is random but subject compliance is imperfect. The bounds provided are the tightest possible, given the distribution of assignments, treatments, and responses. The formulas show that even with high rates of noncompliance, experimental data can yield useful and sometimes accurate information on the average effect of a treatment on the population.

引用

页码：1171 / 1176

页数：6

共 15 条

[1]

Angrist JD, 1996, J AM STAT ASSOC, V91, P444, DOI 10.2307/2291629

[2]

[Anonymous], 1993, R199 U CAL

[3]

BALKE A, 1994, P 10 C UNC ART INT S, P46

[4]

BALKE A, 1995, R242 U CAL COMP SCI

[5]

Chickering DM, 1996, PROCEEDINGS OF THE THIRTEENTH NATIONAL CONFERENCE ON ARTIFICIAL INTELLIGENCE AND THE EIGHTH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE, VOLS 1 AND 2, P1269

[6]

EFRON B, 1991, J AM STAT ASSOC, V86, P9, DOI 10.2307/2289707

[7]

Holland P. W., 1988, ETS Research Report Series, V1988, pi, DOI 10.1002/j.2330-8516.1988.tb00270.x

[8]

*LIP RES CLIN PROG, 1984, JAMA-J AM MED ASSOC, V251, P351, DOI DOI 10.1001/JAMA.251.3.351

[9]

MANSKI CF, 1990, AM ECON REV, V80, P319

[10]

Pearl J, 1995, BIOMETRIKA, V82, P669, DOI 10.1093/biomet/82.4.669

← 1 2 →