When is ATE enough? Risk aversion and inequality aversion in evaluating training programs

Research output: Chapter in Book/Report/Conference proceedingChapter

Abstract

Programs are typically evaluated through the average treatment effect and its standard error. In particular, is the treatment effect positive and is it statistically significant? In theory, programs should be evaluated in a decision framework, using social welfare functions and posterior predictive distributions for outcomes of interest. This chapter discusses the use of stochastic dominance of predictive distributions of outcomes to rank programs, and, under more restrictive parametric and functional form assumptions, the chapter develops intuitive mean-variance tests for program evaluation that are consistent with the underlying decision problem. These concepts are applied to the GAIN and JTPA datasets.

Original languageEnglish (US)
Title of host publicationModelling and Evaluating Treatment Effects in Econometrics
Pages263-287
Number of pages25
Volume21
DOIs
StatePublished - 2007

Publication series

NameAdvances in Econometrics
Volume21
ISSN (Print)07319053

Fingerprint

Risk aversion
Predictive distribution
Inequality aversion
Training program
Stochastic dominance
Program theory
Average treatment effect
Program evaluation
Functional form
Standard error
Treatment effects
Social welfare function
Mean-variance

ASJC Scopus subject areas

  • Economics and Econometrics

Cite this

Dehejia, R. (2007). When is ATE enough? Risk aversion and inequality aversion in evaluating training programs. In Modelling and Evaluating Treatment Effects in Econometrics (Vol. 21, pp. 263-287). (Advances in Econometrics; Vol. 21). https://doi.org/10.1016/S0731-9053(07)00009-6

When is ATE enough? Risk aversion and inequality aversion in evaluating training programs. / Dehejia, Rajeev.

Modelling and Evaluating Treatment Effects in Econometrics. Vol. 21 2007. p. 263-287 (Advances in Econometrics; Vol. 21).

Research output: Chapter in Book/Report/Conference proceedingChapter

Dehejia, R 2007, When is ATE enough? Risk aversion and inequality aversion in evaluating training programs. in Modelling and Evaluating Treatment Effects in Econometrics. vol. 21, Advances in Econometrics, vol. 21, pp. 263-287. https://doi.org/10.1016/S0731-9053(07)00009-6
Dehejia R. When is ATE enough? Risk aversion and inequality aversion in evaluating training programs. In Modelling and Evaluating Treatment Effects in Econometrics. Vol. 21. 2007. p. 263-287. (Advances in Econometrics). https://doi.org/10.1016/S0731-9053(07)00009-6
Dehejia, Rajeev. / When is ATE enough? Risk aversion and inequality aversion in evaluating training programs. Modelling and Evaluating Treatment Effects in Econometrics. Vol. 21 2007. pp. 263-287 (Advances in Econometrics).
@inbook{0b6f9ab27efe44a2a237ac55ca6c111c,
title = "When is ATE enough? Risk aversion and inequality aversion in evaluating training programs",
abstract = "Programs are typically evaluated through the average treatment effect and its standard error. In particular, is the treatment effect positive and is it statistically significant? In theory, programs should be evaluated in a decision framework, using social welfare functions and posterior predictive distributions for outcomes of interest. This chapter discusses the use of stochastic dominance of predictive distributions of outcomes to rank programs, and, under more restrictive parametric and functional form assumptions, the chapter develops intuitive mean-variance tests for program evaluation that are consistent with the underlying decision problem. These concepts are applied to the GAIN and JTPA datasets.",
author = "Rajeev Dehejia",
year = "2007",
doi = "10.1016/S0731-9053(07)00009-6",
language = "English (US)",
isbn = "9780762313808",
volume = "21",
series = "Advances in Econometrics",
pages = "263--287",
booktitle = "Modelling and Evaluating Treatment Effects in Econometrics",

}

TY - CHAP

T1 - When is ATE enough? Risk aversion and inequality aversion in evaluating training programs

AU - Dehejia, Rajeev

PY - 2007

Y1 - 2007

N2 - Programs are typically evaluated through the average treatment effect and its standard error. In particular, is the treatment effect positive and is it statistically significant? In theory, programs should be evaluated in a decision framework, using social welfare functions and posterior predictive distributions for outcomes of interest. This chapter discusses the use of stochastic dominance of predictive distributions of outcomes to rank programs, and, under more restrictive parametric and functional form assumptions, the chapter develops intuitive mean-variance tests for program evaluation that are consistent with the underlying decision problem. These concepts are applied to the GAIN and JTPA datasets.

AB - Programs are typically evaluated through the average treatment effect and its standard error. In particular, is the treatment effect positive and is it statistically significant? In theory, programs should be evaluated in a decision framework, using social welfare functions and posterior predictive distributions for outcomes of interest. This chapter discusses the use of stochastic dominance of predictive distributions of outcomes to rank programs, and, under more restrictive parametric and functional form assumptions, the chapter develops intuitive mean-variance tests for program evaluation that are consistent with the underlying decision problem. These concepts are applied to the GAIN and JTPA datasets.

UR - http://www.scopus.com/inward/record.url?scp=39749086850&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=39749086850&partnerID=8YFLogxK

U2 - 10.1016/S0731-9053(07)00009-6

DO - 10.1016/S0731-9053(07)00009-6

M3 - Chapter

SN - 9780762313808

VL - 21

T3 - Advances in Econometrics

SP - 263

EP - 287

BT - Modelling and Evaluating Treatment Effects in Econometrics

ER -