High-order coverage of smoothed Bayesian bootstrap intervals for population quantiles

Year: 
2019
Working Paper Number: 
WP 19-14
Abstract: 

Using fractional order statistics, we characterize the high-order frequentist coverage probability of smoothed and unsmoothed Bayesian bootstrap credible intervals for population quantiles.  The original Rubin (1981) unsmoothed intervals have O(n-1/2) coverage error, whereas intervals based on the smoothed Bayesian bootstrap of Banks (1988) have much smaller O(n−3/2[log(n)]3) coverage error.  No smoothing parameter is required.  In special cases, the smoothed intervals are exact.  Simulations illustrate our results.

JEL Codes: 
C21
Authors: 

Lonnie Hofmann

PDF: