Bootstrap of Deviation Probabilities with Applications

Ratan Dasgupta

Abstract

We show that under different moment bounds on the underlying variables, bootstrap approximation to the large deviation probabilities of standardized sample sum, based on independent random variables, is valid for a wider zone of n, the sample size, compared to the classical normal tail probability approximation. As an application, different notions of efficiency for statistical tests are considered from Bayesian point of view. In particular, efficiency due to Pitman (1938), Chernoff (1952), and Bayes risk efficiency due to Rubin and Sethuraman (1965) turn out to be special cases with the choice of the weight function; i.e., prior density times loss. The technique is explained by a data example where two processes with different levels of noise are considered. Such a situation is frequently encountered in electronic recordings like EEG/ECG etc., where noise is usually associated with signal.