ProDenICA: Product Density Estimation for ICA using tilted Gaussian density estimates

Share:

A direct and flexible method for estimating an ICA model. This approach estimates the densities for each component directly via a tilted gaussian. The tilt functions are estimated via a GAM poisson model. Details can be found in "Elements of Statistical Learning (2nd Edition)" Section 14.7.4

Author
Trevor Hastie, Rob Tibshirani
Date of publication
2010-05-01 17:45:37
Maintainer
Trevor Hastie <hastie@stanford.edu>
License
GPL-2
Version
1.0
URLs

View on CRAN

Man pages

amari
Compute the 'Amari' distance between two matrices
G1
FastICA contrast functions.
GPois
Fit a tilted Gaussian density via a Poisson GAM
ICAorthW
turn a matrix W into an orthogonal matrix
mixmat
Generate a random mixing matrix with condition number between...
ProDenICA
Product Density Independent Component Analysis
ProDenICA-internal
Internal ProDenICA functions
rjordan
Generate source densities for ICA

Files in this package

ProDenICA
ProDenICA/DESCRIPTION
ProDenICA/man
ProDenICA/man/amari.Rd
ProDenICA/man/G1.Rd
ProDenICA/man/GPois.Rd
ProDenICA/man/ICAorthW.Rd
ProDenICA/man/mixmat.Rd
ProDenICA/man/ProDenICA-internal.Rd
ProDenICA/man/ProDenICA.Rd
ProDenICA/man/rjordan.Rd
ProDenICA/NAMESPACE
ProDenICA/R
ProDenICA/R/amari.R
ProDenICA/R/d.gaussmix.R
ProDenICA/R/djordan.R
ProDenICA/R/dmix.dexp.R
ProDenICA/R/G0.R
ProDenICA/R/G1.R
ProDenICA/R/GPois.R
ProDenICA/R/ICAorthW.R
ProDenICA/R/mixmat.R
ProDenICA/R/modulus.R
ProDenICA/R/plot.ProDenICA.R
ProDenICA/R/ProDenICA.R
ProDenICA/R/r.gaussmix.R
ProDenICA/R/rjordan.R
ProDenICA/R/rmix.dexp.R