Software: Treatment effects estimators (Codes in STATA, R or Gauss)
The Chair of Econometrics has a strong focus on developing new microeconometric treatment effect estimators for impact evaluation. This page intends to make these new estimators available to a wider audience. Usage of these estimators is free of charge but credit should be given to its source, i.e. citation of the original articles. Most of the software is provided as STATA, R or Gauss code. If you use this code and translate it e.g. to Stata or Matlab, please let us know of your innovations.
We show nonparametric identification of quantile treatment effects (QTE) in the regression discontinuity design RDD. The distributional impacts of social programs such as welfare, education or training programs are of large interest to economists. QTE are an intuitive tool to characterize the effects of these interventions on the outcome distribution.
Frandsen, Frölich, Melly (2012): Quantile treatment effects in the regression discontinuity design, Journal of Econometrics, 168 (2), 382–395.
Frölich, Melly (2010): Quantile Treatment Effects in the Regression Discontinuity Design: Process Results and Gini Coefficient, IZA Discussion Paper 4993.
We develop IV estimators for unconditional quantile treatment effects (QTE) when the treatment selection is endogenous, also permitting covariates X. Unconditional QTE summarize the effects of a treatment for the entire (complier) population. They are usually of most interest in policy evaluations because the results can easily be conveyed and summarized. In addition to estimating unconditional QTE of Frölich, Melly (2010, 2013), the command ivqte2 also covers the quantile regression estimator of Koenker and Bassett (1978) extended to heteroskedasticity consistent standard errors, the IV quantile regression estimator of Abadie, Angrist, and Imbens (2002 Econometrica) and the estimator for unconditional QTE of Firpo (2007 Econometrica).
Frölich, Melly (2013): Unconditional quantile treatment effects under endogeneity, Journal of Business & Economic Statistics (JBES), 31:3, 346–357.
Frölich, Melly (2010): Estimation of quantile treatment effects with STATA, Stata Journal, 10 (3), 423–457.
In this paper nonparametric instrumental variable estimation of local average treatment effects (LATE) is extended to incorporate covariates. Including covariates in the estimation of LATE is necessary when the instrumental variable itself is confounded, such that the IV assumptions are valid only conditional on covariates. Previous approaches to handle covariates in the estimation of LATE relied on parametric or semiparametric methods. In this paper, a nonparametric estimator for the estimation of LATE with covariates is developed which is root-n asymptotically normal and efficient.
Frölich (2007): Nonparametric IV estimation of local average treatment effects with covariates, Journal of Econometrics, 139, 35–75.
The command locreg implements local linear and local logit estimators for mixed data (continuous, ordered discrete, unordered discrete and binary regressors). Smoothing parameters are chosen by cross-validation.
Local logit regression for bounded dependent variables; Gauss code
Frölich (2007): Nonparametric regression for binary dependent variables, Econometrics Journal, 9, 511–540.
Gauss code for Frölich (2007) “Finite sample properties of propensity.score matching and weighting estimators, Review of Economics and Statistics, 86(1), 77–90.
Gauss code available here.
Please do not hesitate to contact us in case of questions.