Timothy B. Armstrong

Associate Professor
Department of Economics
University of Southern California

MissAdapt

[ R/Matlab version | Shiny app ]

Implements misspecification adaptive estimators.

Reference:

Armstrong, Timothy B., Patrick Kline, and Liyang Sun. “Adapting to Misspecification” (Aug 2024)

EBCI

[ Stata version | R version | Matlab version ]

Implements robust empirical Bayes confidence intervals.

Reference:

Armstrong, Timothy B., Michal Kolesár, and Mikkel Plagborg-Møller. “Robust Empirical Bayes Confidence Intervals,” with Michal Kolesár and Mikkel Plagborg-Møller, Econometrica, 90, no. 6 (2022): 2567–2602.

RDHonest

[ Stata version | R version ]

Implements honest CIs for regression discontinuity.

References:

Armstrong, Timothy B., and Michal Kolesár. “Simple and Honest Confidence Intervals in Nonparametric Regression,” Quantitative Economics, 2020, 11(1), 1–39

Armstrong, Timothy B., and Michal Kolesár. “Optimal Inference in a Class of Regression Models,” Econometrica, 2018, 86(2), 655–683

GMMSensitivity

[ R version ]

Implements estimators and CIs for sensitivity analysis in moment condition models.

Reference:

Armstrong, Timothy B., and Michal Kolesár.“Sensitivity Analysis using Approximate Moment Condition Models,” with Michal Kolesár, Quantitative Economics, 2021, 12(1), 77–108

ATEHonest

[ R version ]

Implements honest CIs for ATE.

References:

Armstrong, Timothy B., and Michal Kolesár.“Finite-Sample Optimal Estimation and Inference on Average Treatment Effects Under Unconfoundedness,”, with Michal Kolesár, Econometrica, 2021, 89(3), 1141–1177

BWSnooping

[ R version ]

Implements critical values corrected for snooping over bandwidth or propensity score cutoff.

Reference:

Armstrong, Timothy B., and Michal Kolesár. “A simple adjustment for bandwidth snooping,” Review of Economic Studies, 2018, 85(2), 732–65