Parametrization of factor analysis models
From MaRDI portal
The problem of identifiability of factor analysis models representing two blocks of variables is discussed. It is shown that there is a continuum of minimal models which connect two representations of pure regression type. These representations give parametrizations of factor analysis models in terms of which it is possible to solve the problem of identifiability.
Recommendations
Cites work
- Dynamic Factor-Analysis Models for Stationary Processes
- scientific article; zbMATH DE number 3886946 (Why is no real title available?)
- scientific article; zbMATH DE number 3912240 (Why is no real title available?)
- scientific article; zbMATH DE number 3940575 (Why is no real title available?)
- On minimal splitting subspaces and markovian representations
- Realization Theory for Multivariate Stationary Gaussian Processes
- The foundations of factor analysis
- The weak and strong Gaussian probabilistic realization problem
Cited in
(12)- Identifiability of factor analysis: Some results and open problems
- A new method for simultaneous estimation of the factor model parameters, factor scores, and unique parts
- Identifiability of full, marginal, and conditional factor analysis models
- Local identifiability of the factor analysis and measurement error model parameter
- School system evaluation by value added analysis under endogeneity
- Finiteness of small factor analysis models
- On a model of factor analysis with dependant residuals
- scientific article; zbMATH DE number 4097258 (Why is no real title available?)
- scientific article; zbMATH DE number 1136449 (Why is no real title available?)
- Multi‐set factor analysis by means of <scp>P</scp>arafac2
- Algebraic factor analysis: tetrads, pentads and beyond
- A comparison of algorithms for fitting the PARAFAC model
This page was built for publication: Parametrization of factor analysis models
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q916275)
