Benutzer: Gast  Login
Dokumenttyp:
Zeitschriftenaufsatz 
Autor(en):
Leung, Dennis; Drton, Mathias; Hara, Hisayuki 
Titel:
Identifiability of directed Gaussian graphical models with one latent source 
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:
Seitenangaben Beitrag:
394-422 
Sprache:
en 
Volltext / DOI:
Verlag / Institution:
Institute of Mathematical Statistics and Bernoulli Society 
E-ISSN:
1935-7524 
Publikationsdatum:
24.02.2016 
Format:
Text