Algebraic factor analysis: tetrads, pentads and beyond

DOI10.1007/S00440-006-0033-2zbMATH Open1111.13020arXivmath/0509390OpenAlexW3122804471WikidataQ57566451 ScholiaQ57566451MaRDI QIDQ880940FDOQ880940

Authors: Mathias Drton, Bernd Sturmfels, Seth Sullivant 

Publication date: 21 May 2007

Published in: Zeitschrift für Wahrscheinlichkeitstheorie und Verwandte Gebiete (Search for Journal in Brave)

Abstract: Factor analysis refers to a statistical model in which observed variables are conditionally independent given fewer hidden variables, known as factors, and all the random variables follow a multivariate normal distribution. The parameter space of a factor analysis model is a subset of the cone of positive definite matrices. This parameter space is studied from the perspective of computational algebraic geometry. Gr"obner bases and resultants are applied to compute the ideal of all polynomial functions that vanish on the parameter space. These polynomials, known as model invariants, arise from rank conditions on a symmetric matrix under elimination of the diagonal entries of the matrix. Besides revealing the geometry of the factor analysis model, the model invariants also furnish useful statistics for testing goodness-of-fit.

Full work available at URL: https://arxiv.org/abs/math/0509390

Recommendations

• The algebra of multimode factor analysis
• Parametrization of factor analysis models
• Lectures on algebraic statistics
• Instability of statistical factor analysis
• Algebraic statistical models

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05) Knowledge representation (68T30) Gröbner bases; other bases for ideals and modules (e.g., Janet and border bases) (13P10) Projective techniques in algebraic geometry (14N05)

Cites Work

• SINGULAR
• Title not available (Why is that?)
• Probability for Statisticians
• Title not available (Why is that?)
• The rank of reduced dispersion matrices
• Title not available (Why is that?)
• Title not available (Why is that?)
• Algebraic Statistics for Computational Biology
• EM algorithms for ML factor analysis
• Title not available (Why is that?)
• Title not available (Why is that?)
• Algebraic analysis for nonidentifiable learning machines
• Title not available (Why is that?)
• Higher Lawrence configurations.
• Combinatorial secant varieties
• The hyperoctahedral group, symmetric group representations and the moments of the real Wishart distribution
• On some models for multivariate binary variables parallel in complexity with the multivariate Gaussian distribution
• Stratified exponential families: Graphical models and model selection
• Phylogenetic ideals and varieties for the general Markov model
• The maximum likelihood degree
• Conditionals, Information, and Inference
• Gröbner bases of ideals of minors of a symmetric matrix
• The Isserlis matrix and its application to non-decomposable graphical Gaussian models
• MacMahon's master theorem, representation theory, and moments of Wishart distributions
• Title not available (Why is that?)
• Gröbner bases and triangulations of the second hypersimplex
• Title not available (Why is that?)
• Instability of statistical factor analysis
• On identification of multi-factor models with correlated residuals

Cited In (30)

• Equivariant Gröbner bases and the Gaussian two-factor model
• Finiteness of small factor analysis models
• Half-trek criterion for identifiability of latent variable models
• Set-theoretic defining equations of the tangential variety of the Segre variety
• Order-invariant prior specification in Bayesian factor analysis
• Moments of minors of Wishart matrices
• Instability of statistical factor analysis
• Initial ideals of Pfaffian ideals
• Three notions of tropical rank for symmetric matrices
• Algebraic structures in statistical methodology. Abstracts from the workshop held December 4--10, 2022
• A note on one-factor analysis
• A Gröbner basis for the secant ideal of the second hypersimplex
• Structurally determined inequality constraints on correlations in the cycle of linear dependencies
• Exploring the total positivity of yields correlations
• Complex best \(r\)-term approximations almost always exist in finite dimensions
• Finite Gröbner bases in infinite dimensional polynomial rings and applications
• Moment Varieties for Mixtures of Products
• Wald tests of singular hypotheses
• Qualitative inequalities for squared partial correlations of a Gaussian random vector
• Dimensions of higher order factor analysis models
• Gaussian graphical models with toric vanishing ideals
• Algebraic geometry of Gaussian Bayesian networks
• Minimal generators for symmetric ideals
• Identifiability of directed Gaussian graphical models with one latent source
• Finite generation of symmetric ideals
• Finiteness for the \(k\)-factor model and chirality varieties
• Title not available (Why is that?)
• Likelihood ratio tests and singularities
• Nested covariance determinants and restricted trek separation in Gaussian graphical models
• Singularity Structures and Impacts on Parameter Estimation in Finite Mixtures of Distributions

Uses Software

• Macaulay2
• SINGULAR
• TETRAD

This page was built for publication: Algebraic factor analysis: tetrads, pentads and beyond

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q880940)