Answers are grounded in this paper and cite their sources.
We present a general approach to construct confidence sets for real-valued and multidimensional parameters for a broad range of randomization schemes by inverting Monte Carlo tests. The approach exploits two facts: (i) there are Monte Carlo tests that are conservative despite relying on simulation, and (ii) since the coverage probability of confidence sets depends only on the significance level of the test of the true null, every null can be tested using the same Monte Carlo sample. When the parameter is real-valued and the p-value is quasiconcave in the parameter with the data and Monte Carlo sample held fixed, conservative confidence sets can be constructed in O(n) time, where n is the number of data. The values of some test statistics for different Monte Carlo samples and parameter values have a simple relationship that makes more savings possible. We provide open-source Python and R implementations, available online.
No takes yet. Share an insight, caveat, or question.
Glazer et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: