On adaptive inference and confidence bands

Marc Hoffmann, Richard Nickl
2011 Annals of Statistics  
The problem of existence of adaptive confidence bands for an unknown density $f$ that belongs to a nested scale of H\"{o}lder classes over $\mathbb{R}$ or $[0,1]$ is considered. Whereas honest adaptive inference in this problem is impossible already for a pair of H\"{o}lder balls $\Sigma(r),\Sigma(s),r\ne s$, of fixed radius, a nonparametric distinguishability condition is introduced under which adaptive confidence bands can be shown to exist. It is further shown that this condition is
more » ... ndition is necessary and sufficient for the existence of honest asymptotic confidence bands, and that it is strictly weaker than similar analytic conditions recently employed in Gin\'{e} and Nickl [Ann. Statist. 38 (2010) 1122--1170]. The exceptional sets for which honest inference is not possible have vanishingly small probability under natural priors on H\"{o}lder balls $\Sigma(s)$. If no upper bound for the radius of the H\"{o}lder balls is known, a price for adaptation has to be paid, and near-optimal adaptation is possible for standard procedures. The implications of these findings for a general theory of adaptive inference are discussed.
doi:10.1214/11-aos903 fatcat:cmk3cd27sfhtdn3zg67fz5vxdu