Estimation of the Multi-Dimensional Probability Density Function.
Abstract
Suppose X1,X2,... are independent, identically distributed (i.i.d.) (R sup d) valued random variables with common probability density function f. A problem of considerable practical importance and also of theoretical interest is the estimation of f through some statistic based on the observed sequence (X sub k). Such statistics are called empirical density functions, and the author examines the rate of convergence of the empirical densities to f. The author also considers the situation when there is 'noise' in the observations (X sub k).
Document Details
- Document Type
- Technical Report
- Publication Date
- Jul 01, 1976
- Accession Number
- ADA029965
Entities
People
- J. Kuelbs
Organizations
- University of Wisconsin–Madison