Dear all,

Thanks to Prof Baum, a new command is now available on SSC. This new command, named -ivbounds-, provides an instrumental variable estimate of the bounds of the average causal effect for compliers (Imbens and Angrist, 1994) when both noncompliance and misreporting of treatment status are present. The approach follows the estimation procedure developed in Tommasi and Zhang (2020) (https://drive.google.com/file/d/140o...sRj7r8LuJ/view).

Misreporting (misclassification) of the treatment status due to the desire to shorten the time spent on the interview, stigma of program participation, or sensitivity of income information, is an increasing problem for social scientists. -ivbounds- is the only Stata command, together with -tebounds-, that allows the practitioner to deal with all these issues.

-ivbounds- assumes a binary treatment variable, a binary or discrete instrumental variable, and a continuous or discrete dependent variable. The bounds can be improved if external information regarding the treatment misclassification probability in the treatment status is available.

When the instrumental variable is binary, the command estimates the bounds of the local average treatment effect (LATE). When the instrumental variable is binary and covariates are included, the command estimates the bounds of the unconditional LATE (Frölich, 2007). In both cases, the estimated bounds coincide with those in Ura (2018).

When the instrumental variable is discrete, the command estimates the bounds of the weighted average of LATEs (WLATE). When the instrumental variable is discrete and covariates are included, the command estimates the bounds of the conditional WLATE. In the latter case, the user must specify the number of strata to discretise the propensity score. The inference procedure follows Dehejia and Wahba (1999) and Battistin and Sianesi (2011).

This command requires -moremata-, -bsweights-, and -bs4rw- to run. So make sure they are installed before trying the examples in the helpfile or the do file that comes with the command.

Please feel free to share your comments, reports of bugs, and propositions for extensions. I hope this will be useful.

Thank you,

Denni