Benutzer: Gast  Login
Titel:

Identifiability of directed Gaussian graphical models with one latent source

Dokumenttyp:
Zeitschriftenaufsatz
Autor(en):
Leung, Dennis; Drton, Mathias; Hara, Hisayuki
Abstract:
We study parameter identifiability of directed Gaussian graphical models with one latent variable. In the scenario we consider, the latent variable is a confounder that forms a source node of the graph and is a parent to all other nodes, which correspond to the observed variables. We give a graphical condition that is sufficient for the Jacobian matrix of the parametrization map to be full rank, which entails that the parametrization is generically finite-to-one, a fact that is sometimes also re...     »
Stichworte:
Covariance matrix, factor analysis, graphical model, parameter identification, structural equation model
Dewey Dezimalklassifikation:
510 Mathematik
Zeitschriftentitel:
Electronic Journal of Statistics
Jahr:
2016
Band / Volume:
10
Jahr / Monat:
2016-02
Quartal:
1. Quartal
Monat:
Feb
Heft / Issue:
1
Seitenangaben Beitrag:
394-422
Sprache:
en
Volltext / DOI:
doi:10.1214/16-ejs1111
WWW:
Project Euclid
Verlag / Institution:
Institute of Mathematical Statistics and Bernoulli Society
E-ISSN:
1935-7524
Publikationsdatum:
24.02.2016
Format:
Text
 BibTeX