Data-driven kernel representations for sampling with an unknown block dependence structure under correlation constraints

Abstract : The multidimensional Gaussian kernel-density estimation (G-KDE) is a powerful tool to identify the distribution of random vectors when the maximal information is a set of independent realizations. For these methods, a key issue is the choice of the kernel and the optimization of the bandwidth matrix. To optimize these kernel representations, two adaptations of the classical G-KDE are presented. First, it is proposed to add constraints on the mean and the covariance matrix in the G-KDE formalism. Secondly, it is suggested to separate in different groups the components of the random vector of interest that could reasonably be considered as independent. This block by block decomposition is carried out by looking for the maximum of a cross-validation likelihood quantity that is associated with the block formation. This leads to a tensorized version of the classical G-KDE. Finally, it is shown on a series of examples how these two adaptations can improve the nonparametric representations of the densities of random vectors, especially when the number of available realizations is relatively low compared to their dimensions.
Complete list of metadatas

https://hal-upec-upem.archives-ouvertes.fr/hal-01794809
Contributor : Guillaume Perrin <>
Submitted on : Thursday, May 17, 2018 - 8:33:03 PM
Last modification on : Thursday, July 18, 2019 - 4:36:07 PM
Long-term archiving on : Tuesday, September 25, 2018 - 5:27:53 PM

File

Data_driven_PERRIN_JMS.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01794809, version 1

Collections

Citation

Guillaume Perrin, C Soize, N. Ouhbi. Data-driven kernel representations for sampling with an unknown block dependence structure under correlation constraints. Computational Statistics and Data Analysis, Elsevier, 2018. ⟨hal-01794809⟩

Share

Metrics

Record views

126

Files downloads

116