DeepAI AI Chat
Log In Sign Up

Perturbed factor analysis: Improving generalizability across studies

by   Arkaprava Roy, et al.
Duke University

Factor analysis is routinely used for dimensionality reduction. However, a major issue is `brittleness' in which one can obtain substantially different factors in analyzing similar datasets. Factor models have been developed for multi-study data by using additive expansions incorporating common and study-specific factors. However, allowing study-specific factors runs counter to the goal of producing a single set of factors that hold across studies. As an alternative, we propose a class of Perturbed Factor Analysis (PFA) models that assume a common factor structure across studies after perturbing the data via multiplication by a study-specific matrix. Bayesian inference algorithms can be easily modified in this case by using a matrix normal hierarchical model for the perturbation matrices. The resulting model is just as flexible as current approaches in allowing arbitrarily large differences across studies, but has substantial advantages that we illustrate in simulation studies and an application to NHANES data. We additionally show advantages of PFA in single study data analyses in which we assign each individual their own perturbation matrix, including reduced generalization error and improved identifiability.


page 5

page 12

page 14

page 15

page 18

page 19

page 20

page 22


Bayesian Combinatorial Multi-Study Factor Analysis

Analyzing multiple studies allows leveraging data from a range of source...

Bayesian factor models for multivariate categorical data obtained from questionnaires

Factor analysis is a flexible technique for assessment of multivariate d...

Bayesian Multi-study Factor Analysis for High-throughput Biological Data

This paper presents a new modeling strategy for joint unsupervised analy...

On the Dimensional Indeterminacy of One-Wave Factor Analysis Under Causal Effects

It is shown, with two sets of survey items that separately load on two d...

R-factor analysis of data generated by a combination of R- and Q-factors leads to biased loading estimates

The effect of combined, generating R- and Q-factors of measured variable...

Score Predictor Factor Analysis as model for the identification of single-item indicators

Score Predictor Factor Analysis (SPFA) was introduced as a method that a...

A Bayesian Approach to Spherical Factor Analysis for Binary Data

Factor models are widely used across diverse areas of application for pu...