User: Guest  Login
Title:

Half-trek criterion for identifiability of latent variable models

Document type:
Zeitschriftenaufsatz
Author(s):
Barber, Rina Foygel; Drton, Mathias; Sturma, Nils; Weihs, Luca
Abstract:
We consider linear structural equation models with latent variables and develop a criterion to certify whether the direct causal effects between the observable variables are identifiable based on the observed covariance matrix. Linear structural equation models assume that both observed and latent variables solve a linear equation system featuring stochastic noise terms. Each model corresponds to a directed graph whose edges represent the direct effects that appear as coefficients in the equatio...     »
Keywords:
Covariance matrix, factor analysis, Graphical model, hidden variables, latent variables, parameter identification, structural equation model
Dewey Decimal Classification:
510 Mathematik
Journal title:
The Annals of Statistics
Year:
2022
Journal volume:
50
Year / month:
2022-12
Quarter:
4. Quartal
Month:
Dec
Journal issue:
6
Language:
en
Fulltext / DOI:
doi:10.1214/22-aos2221
WWW:
PROJECTeuclid
Publisher:
Institute of Mathematical Statistics
E-ISSN:
0090-5364
Date of publication:
01.12.2022
TUM Institution:
Lehrstuhl für Mathematische Statistik
Format:
Text
 BibTeX