Abstract
This paper studies analytic inference with several dimensions of clustering. In such setups, the commonly used approach has two drawbacks. First, the corresponding variance estimator is not necessarily positive. Second, inference is invalid in non-Gaussian regimes, namely when the estimator of the parameter of interest is not asymptotically Gaussian. We consider a simple fix of the usual approach that addresses both issues. In Gaussian regimes, inference is asymptotically exact and equivalent to the usual one. Otherwise, inference is asymptotically conservative. Inference is also uniformly valid over a certain class of data generating processes. Finally, we highlight potential issues with tests of multiple conditions or inference based on nonlinear estimators in this setup.
Local Organizer: Giovanni Angelini