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 language | English (US) |
---|---|
Pages (from-to) | 619-635 |
Number of pages | 17 |
Journal | Journal of the Royal Statistical Society. Series C: Applied Statistics |
Volume | 61 |
Issue number | 4 |
DOIs | |
State | Published - Aug 2012 |
Externally published | Yes |
Keywords
- Covariance estimation
- Factor models
- Multivariate probit models
- Voting patterns
ASJC Scopus subject areas
- Statistics and Probability
- Statistics, Probability and Uncertainty