Let ÿn(t) denote the weighted (smooth) bootstrap process of an empirical process. We show that the order of the best Gaussian approximation for ÿn(t) is n -1=2 log n and we construct a sequence of approximating Brownian bridges achieving this rate. We also obtain an approximation for ÿn(t) using a s
On the best approximation for bootstrapped empirical processes
✍ Scribed by Lajos Horváth; Josef Steinebach
- Publisher
- Elsevier Science
- Year
- 1999
- Tongue
- English
- Weight
- 250 KB
- Volume
- 41
- Category
- Article
- ISSN
- 0167-7152
No coin nor oath required. For personal study only.
✦ Synopsis
We obtain a lower bound for the rate of approximation of bootstrapped empirical processes with Brownian bridges.
📜 SIMILAR VOLUMES
The total variation distance is estimated between distributions of the so-called rescaled empirical process and a Poisson point process which are indexed by the all Borel subsets of a bounded Borel set in R k :
An explicit proof is given that no sequence of quasi-classical adiabatic approximations in which intrinsic symmetry re. strictions are relased in any way can give an approprhtely approximated partition function sm;Fller than the one diRctly obtainable in the absence of such a sequence.