Mhbounds - Sensitivity Analysis for Average Treatment Effects

By Sascha O. Becker, Marco Caliendo. Social Science Research Network, 2007

169 citations recorded by the source corpus.

Matching has become a popular approach to estimate average treatment effects. It is based on the conditional independence or unconfoundedness assumption. Checking the sensitivity of the estimated results with respect to deviations from this identifying assumption has become an increasingly important topic in the applied evaluation literature. If there are unobserved variables which affect assignment into treatment and the outcome variable simultaneously, a hidden bias might arise to which matching estimators are not robust. We address this problem with the bounding approach proposed by Rosenbaum (2002), where mhbounds allows the researcher to determine how strongly an unmeasured variable must influence the selection process in order to undermine the implications of the matching analysis.

Subfields: Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/4a7117fb199fe5bd5de325a0fbd821da26b0ecd2

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Statistics and Probability

Citations in the corpus

Start-Up Subsidies for the Unemployed: Long-Term Evidence and Effect Heterogeneity (2011, background): “The 16 underlying test statistic then tests up to which extent this unobserved factor Γ will influence the significance of the results (see Becker and Caliendo, 2007, for more details on the implementation of the test procedure and the STATA module mhbounds.ado).”
Food standards, certification, and poverty among coffee farmers in Uganda (2015, methodology): “To test the robustness of our results we calculate Rosenbaum bounds for hidden bias (DiPrete & Gangl, 2004; Becker & Caliendo, 2007).”
Got milk? The impact of Heifer International's livestock donation programs in Rwanda on nutritional outcomes (2014, background): “…of unobserved characteristics that influence both assignment into treatment and the outcome (Rosenbaum, 2002), we estimate Rosenbaum bounds to determine the extent to which the significance of the results relies on the untestable conditional independence assumption (Becker and Caliendo, 2007).”
Parental Incarceration, Child Homelessness, and the Invisible Consequences of Mass Imprisonment (2014, methodology): “…driving the results herein. to quantify how large such an unobserved factor would have to be to undermine my results, I present Mantel-Haenszel bounds in table 4 for each propensity score model that produced significant results using StAtA-compatible software designed by Becker and Caliendo (2007).”