Event Date
SPEAKER: Ryan Giordano / Post-Doctoral Researcher, Massachusetts Institute of Technology
TITLE: "An Automatic Finite-Sample Robustness Metric: Can Dropping a Little Data Make a Big Difference?"
ABSTRACT:
I propose a method to assess the sensitivity of statistical analyses to the
removal of a small fraction of the data. Manually checking the influence of all
possible small subsets is computationally infeasible, so I provide an
approximation to find the most influential subset. My metric, the ``Approximate
Maximum Influence Perturbation,'' is automatically computable for common methods
including (but not limited to) OLS, IV, MLE, GMM, and variational Bayes. I
provide finite-sample error bounds on the approximation performance and, at
minimal extra cost, I provide an exact finite-sample lower bound on sensitivity.
I find that sensitivity is driven by a signal-to-noise ratio in the inference
problem, is not reflected in standard errors, does not disappear asymptotically,
and is not due to misspecification. While some empirical applications are
robust, results of several economics papers can be overturned by removing less
than 1% of the sample.
This seminar will be presented remotely via Zoom.