# Analyse canonique généralisée d'un flux de données d'espérance variable dans le temps

1 BIGS - Biology, genetics and statistics
Inria Nancy - Grand Est, IECL - Institut Élie Cartan de Lorraine
Abstract : Consider a data stream and suppose that each multidimensional data $z_n$ is a realization of a random vector $Z_n$ whose expectation $\theta_n$ varies with time. Let $\tilde{Z}_n=Z_n-\theta_n$ and suppose that the vectors $\tilde{Z}_n$ form an i.i.d. sample of a random vector $\tilde{Z}$. Stochastic approximation processes using data blocks are used to estimate on-line direction vectors of the principal axes of the generalized canonical correlation analysis of $\tilde{Z}$.
keyword :
Document type :
Conference papers
Domain :

Cited literature [5 references]

https://hal.inria.fr/hal-00750883
Contributor : Romain Bar Connect in order to contact the contributor
Submitted on : Tuesday, November 13, 2012 - 9:28:24 AM
Last modification on : Thursday, January 20, 2022 - 5:27:04 PM
Long-term archiving on: : Thursday, February 14, 2013 - 3:05:09 AM

### File

soumission.pdf
Files produced by the author(s)

### Identifiers

• HAL Id : hal-00750883, version 1

### Citation

Jean-Marie Monnez, Romain Bar. Analyse canonique généralisée d'un flux de données d'espérance variable dans le temps. XIXèmes Rencontres de la Société Francophone de Classification - SFC 2012, Oct 2012, Marseille, France. ⟨hal-00750883⟩

Record views