cfcompare: Compare DID, SDID, MC and the Triply Robust Panel estimator

cfcompare lets applied researchers run several panel causal-inference estimators on the same data and compare them on one tidy schema and one plot. It provides a native R implementation of the Triply RObust Panel (TROP) estimator of Athey, Imbens, Qu & Viviano (2026), and a comparison harness that also wraps difference-in-differences (TWFE), matrix completion, and – via the synthdid package – synthetic difference-in-differences and synthetic control.

Details

Key functions:

• trop() – the native TROP estimator.

• panel_compare() – run DID / SDID / MC / TROP together.

• as_att() – coerce external results into the shared schema.

• autoplot.cf_comparison(), plot_counterfactual() – visuals.

• sim_panel() – simulate factor-model panels for experiments.

References

Athey, S., Imbens, G. W., Qu, Z., & Viviano, D. (2026). Triply Robust Panel Estimators. Journal of Applied Econometrics. doi:10.1002/jae.70061 .

See also

Useful links:

• https://github.com/takuma1102/cfcompare

• https://takuma1102.github.io/cfcompare/

• Report bugs at https://github.com/takuma1102/cfcompare/issues

Author

Maintainer: Takuma Iwasaki iwasakit@stanford.edu [copyright holder]

Authors:

• Takuma Iwasaki iwasakit@stanford.edu [copyright holder]