A sparse factor analytic probit model for congressional voting patterns

Paul Hahn, Carlos M. Carvalho, James G. Scott

Research output: Contribution to journalArticle

14 Citations (Scopus)

Abstract

The paper adapts sparse factor models for exploring covariation in multivariate binary data, with an application to measuring latent factors in US Congressional roll-call voting patterns. This straightforward modification provides two advantages over traditional factor analysis of binary data. First, a sparsity prior can be used to assess the evidence that a given factor loading may be exactly 0, realizing a principled unification of exploratory and confirmatory factory analysis. Second, incorporating sparsity into existing factor analytic probit models effects a favourable bias-variance trade-off in estimating the covariance matrix of the multivariate Gaussian latent variables. Posterior summaries from this model applied to the roll-call data provide novel metrics of partisanship of a given Senate.

Original languageEnglish (US)
Pages (from-to)619-635
Number of pages17
JournalJournal of the Royal Statistical Society. Series C: Applied Statistics
Volume61
Issue number4
DOIs
StatePublished - Aug 1 2012
Externally publishedYes

Fingerprint

Probit Model
Voting
Binary Data
Sparsity
Factor Models
Multivariate Data
Latent Variables
Factor Analysis
Unification
Covariance matrix
Trade-offs
Metric
Factors
Probit model
Model
Latent factors
Latent variables
Factor loadings
Factor analysis
Partisanship

Keywords

  • Covariance estimation
  • Factor models
  • Multivariate probit models
  • Voting patterns

ASJC Scopus subject areas

  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Cite this

A sparse factor analytic probit model for congressional voting patterns. / Hahn, Paul; Carvalho, Carlos M.; Scott, James G.

In: Journal of the Royal Statistical Society. Series C: Applied Statistics, Vol. 61, No. 4, 01.08.2012, p. 619-635.

Research output: Contribution to journalArticle

@article{0588ae45861f40d3aa393420525aa98e,
title = "A sparse factor analytic probit model for congressional voting patterns",
abstract = "The paper adapts sparse factor models for exploring covariation in multivariate binary data, with an application to measuring latent factors in US Congressional roll-call voting patterns. This straightforward modification provides two advantages over traditional factor analysis of binary data. First, a sparsity prior can be used to assess the evidence that a given factor loading may be exactly 0, realizing a principled unification of exploratory and confirmatory factory analysis. Second, incorporating sparsity into existing factor analytic probit models effects a favourable bias-variance trade-off in estimating the covariance matrix of the multivariate Gaussian latent variables. Posterior summaries from this model applied to the roll-call data provide novel metrics of partisanship of a given Senate.",
keywords = "Covariance estimation, Factor models, Multivariate probit models, Voting patterns",
author = "Paul Hahn and Carvalho, {Carlos M.} and Scott, {James G.}",
year = "2012",
month = "8",
day = "1",
doi = "10.1111/j.1467-9876.2012.01044.x",
language = "English (US)",
volume = "61",
pages = "619--635",
journal = "Journal of the Royal Statistical Society. Series C: Applied Statistics",
issn = "0035-9254",
publisher = "Wiley-Blackwell",
number = "4",

}

TY - JOUR

T1 - A sparse factor analytic probit model for congressional voting patterns

AU - Hahn, Paul

AU - Carvalho, Carlos M.

AU - Scott, James G.

PY - 2012/8/1

Y1 - 2012/8/1

N2 - The paper adapts sparse factor models for exploring covariation in multivariate binary data, with an application to measuring latent factors in US Congressional roll-call voting patterns. This straightforward modification provides two advantages over traditional factor analysis of binary data. First, a sparsity prior can be used to assess the evidence that a given factor loading may be exactly 0, realizing a principled unification of exploratory and confirmatory factory analysis. Second, incorporating sparsity into existing factor analytic probit models effects a favourable bias-variance trade-off in estimating the covariance matrix of the multivariate Gaussian latent variables. Posterior summaries from this model applied to the roll-call data provide novel metrics of partisanship of a given Senate.

AB - The paper adapts sparse factor models for exploring covariation in multivariate binary data, with an application to measuring latent factors in US Congressional roll-call voting patterns. This straightforward modification provides two advantages over traditional factor analysis of binary data. First, a sparsity prior can be used to assess the evidence that a given factor loading may be exactly 0, realizing a principled unification of exploratory and confirmatory factory analysis. Second, incorporating sparsity into existing factor analytic probit models effects a favourable bias-variance trade-off in estimating the covariance matrix of the multivariate Gaussian latent variables. Posterior summaries from this model applied to the roll-call data provide novel metrics of partisanship of a given Senate.

KW - Covariance estimation

KW - Factor models

KW - Multivariate probit models

KW - Voting patterns

UR - http://www.scopus.com/inward/record.url?scp=84864101718&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84864101718&partnerID=8YFLogxK

U2 - 10.1111/j.1467-9876.2012.01044.x

DO - 10.1111/j.1467-9876.2012.01044.x

M3 - Article

AN - SCOPUS:84864101718

VL - 61

SP - 619

EP - 635

JO - Journal of the Royal Statistical Society. Series C: Applied Statistics

JF - Journal of the Royal Statistical Society. Series C: Applied Statistics

SN - 0035-9254

IS - 4

ER -