fdid – User Manual

Welcome

This Quarto book serves as a user tutorial for the fdid package in R, which implements the Factorial Difference-in-Differences (FDID) framework introduced by Xu et al. (2026)[Paper]. This design is tailored for panel data settings where all units are exposed to a universal event but vary in a baseline factor \(G\) (currently supported as binary).

fdid provides two main functions:

• fdid_prepare(): reshapes long format panel data into a wide fdid()-ready format.
• fdid(): performs factorial difference-in-differences estimation using several methods.

And two S3 methods:

• summary(): summarizes the estimation results.
• plot(): visualizes raw means, dynamic effects, and model comparisons.

Organization

The user tutorial is structured into the following chapters:

• Chapter 1  Get Started
  Installation instructions and dataset description.

• Chapter 2  Binary G
  Data preparation, estimation with binary \(G\), and basic plots.

• Chapter 3  Continuous G
  Estimation with continuous \(G\) using linear methods.

• Chapter 4  Visualization Options
  Extended visualization options and method comparisons.

• Chapter 5  Sensitivity Analysis
  Sensitivity analysis for unmeasured confounding using sensemakr.

Contributors

• Rivka Lipkovitz (MIT)
  
• Enhan Liu (UChicago Harris)
• Yiqing Xu (Stanford)
  

Report bugs

Please report any bugs to me (yiqingxu [at] stanford.edu) or submit an issue on GitHub. Please include your minimally replicable code and data files and a panelView treatment status plot. Your feedback is highly valued!

fect:

Cao, Jiarui, Yiqing Xu, and Chuanchuan Zhang. 2022. “Clans and Calamity: How Social Capital Saved Lives During China’s Great Famine.” Journal of Development Economics 157: 102865.

Cinelli, Carlos, and Chad Hazlett. 2020. “Making Sense of Sensitivity: Extending Omitted Variable Bias.” Journal of the Royal Statistical Society: Series B 82 (1): 39–67.

Xu, Yiqing, Anqi Zhao, and Peng Ding. 2026. “Factorial Difference-in-Differences.” Journal of the American Statistical Association, nos. just-accepted: 1–23.