Hello Statalisters! I would like to announce that drlate, a Stata module for doubly robust estimation of the local average treatment effect (LATE) and the local average treatment effect on the treated (LATT), is now available in SSC.
drlate provides estimators of the local average treatment effect (LATE) and the local average treatment effect on the treated (LATT), including inverse-probability-weighted regression adjustment (IPWRA), inverse probability weighting (IPW), augmented inverse probability weighting (AIPW), and regression adjustment (RA). Outcome and treatment models can be specified as linear, logistic, or Poisson regressions. The instrument propensity score is specified as a logistic regression (logit) and can be estimated by maximum likelihood (ML), covariate balancing propensity score (CBPS), or inverse probability tilting (IPT). The instrument must be binary.
The estimators implemented by drlate are described in the paper by Słoczyński, Uysal, and Wooldridge, "Doubly Robust Estimation of Local Average Treatment Effects Using Inverse Probability Weighted Regression Adjustment," available here: https://arxiv.org/abs/2208.01300. We expect to have a new draft of this paper in the next few months.
"Dr. Late" can provide a cure when the treatment is confounded, you want to allow heterogeneity, and you want to take functional form seriously!
Let us know if you have comments or suggestions about the package.
drlate provides estimators of the local average treatment effect (LATE) and the local average treatment effect on the treated (LATT), including inverse-probability-weighted regression adjustment (IPWRA), inverse probability weighting (IPW), augmented inverse probability weighting (AIPW), and regression adjustment (RA). Outcome and treatment models can be specified as linear, logistic, or Poisson regressions. The instrument propensity score is specified as a logistic regression (logit) and can be estimated by maximum likelihood (ML), covariate balancing propensity score (CBPS), or inverse probability tilting (IPT). The instrument must be binary.
The estimators implemented by drlate are described in the paper by Słoczyński, Uysal, and Wooldridge, "Doubly Robust Estimation of Local Average Treatment Effects Using Inverse Probability Weighted Regression Adjustment," available here: https://arxiv.org/abs/2208.01300. We expect to have a new draft of this paper in the next few months.
"Dr. Late" can provide a cure when the treatment is confounded, you want to allow heterogeneity, and you want to take functional form seriously!
Let us know if you have comments or suggestions about the package.
