Confidently Comparing Estimators with the c-value

02/19/2021
by   Brian L Trippe, et al.
0

Modern statistics provides an ever-expanding toolkit for estimating unknown parameters. Consequently, applied statisticians frequently face a difficult decision: retain a parameter estimate from a familiar method or replace it with an estimate from a newer or complex one. While it is traditional to compare estimators using risk, such comparisons are rarely conclusive in realistic settings. In response, we propose the "c-value" as a measure of confidence that a new estimate achieves smaller loss than an old estimate on a given dataset. We show that it is unlikely that a computed c-value is large and that the new estimate has larger loss than the old. Therefore, just as a small p-value provides evidence to reject a null hypothesis, a large c-value provides evidence to use a new estimate in place of the old. For a wide class of problems and estimators, we show how to compute a c-value by first constructing a data-dependent high-probability lower bound on the difference in loss. The c-value is frequentist in nature, but we show that it can provide a validation of Bayesian estimates in real data applications involving hierarchical models and Gaussian processes.

READ FULL TEXT

Please sign up or login with your details

Forgot password? Click here to reset