A Variation on Scheffe's Theorem, with Application to Nonparametric Density Estimation.
Abstract
For probability density functions (f sub n) and f defined on a d-dimensional Euclidean space, Scheffe proved the useful result that pointwise convergence of f sub n to f implies convergence in the mean as well. In some applications it is of interest to know the rate of the mean convergence, and, in particular, to connect it with the rate of the uniform pointwise convergence. A basic lemma of this type is derived and applied to some problems in nonparametric density estimation. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- May 01, 1979
- Accession Number
- ADA072132
Entities
People
- Robert Serfling
Organizations
- Florida State University