Cover Image
QR Code

Nearly Nonparametric Multivariate Density Estimates that Incorporate Marginal Parametric Density Information.

A method for modeling multivariate data with fixed marginals that preserves a rich and flexible multivariate structure is presented. Nonparametric estimators that are almost equal to the MLE estimates for the marginal densities and remain close to the kernel nonparametric density estimates for the...

Full description

Published in:American Statistician Vol. 57; no. 3; pp. 183 - 189
Main Authors: Spiegelman, Clifford, Park, Eun Sug
Format: Article
Published: American Statistical Association August 2003
Subjects:
Density functionals
Multivariate analysis
Test validity
Online Access:View this record in EBSCOhost
Summary:A method for modeling multivariate data with fixed marginals that preserves a rich and flexible multivariate structure is presented. Nonparametric estimators that are almost equal to the MLE estimates for the marginal densities and remain close to the kernel nonparametric density estimates for the joint density estimates, given that the assumption concerning the marginal densities is correct, are demonstrated.