Nonparametric and Semiparametric Methods

Software packages for nonparametric and semiparametric smoothing methods with application to causal inference, treatment effect and program evaluation estimation and inference. Replication files and illustration codes employing these packages are also available.

This work was supported in part by the National Science Foundation through grants SES-1459931, SES-1459967, SES-1947662, SES-1947805, and SES-2019432, and by the National Institutes of Health through grant R01 GM072611-16.

Software available in Python, R and/or Stata

• binsreg: partition selection, point estimation, pointwise and uniform inference, and graphical procedures using binscatter methods.
• portsort: estimation and inference using portfolio sorting methods.

• lspartition: estimation and inference using partitioning-based least squares methods, including B-splines, wavelet and piecewise polynomial regression estimators.

• nprobust: estimation and inference using kernel density and local polynomial regression methods.
• lpdensity: estimation and inference using local polynomial distribution/density regression methods.

• lpcde: estimation and inference using local polynomial conditional distribution/density regression methods.

• scpi: estimation and inference using synthetic control methods.

Replication

Replication files and illustration code are available in the replication page.

Contributors

• Sebastian Calonico, UC Davis.
• Matias D. Cattaneo, Princeton University.
• Rajita Chandak, University of Wisconsin-Madison.
• Richard K. Crump, Federal Reserve Bank of New York.
• Max H. Farrell, UC Santa Barbara.
• Yingjie Feng, Tsinghua University.
• Michael Jansson, UC Berkeley and CREATES.
• Xinwei Ma, UC San Diego.
• Ricardo Masini, UC Davis.
• Filippo Palomba, Princeton University.
• Rocio Titiunik, Princeton University.
• Weining Wang, University of York.
• Zhijiang (Tony) Ye, L.E.K. Consulting.