Skip to contentScholarGate
LibraryBookshelfDeskReview StudioAssistant
Sign in
On this page
IntuitionHow it worksWhen to use itStrengths & limitationsCommon pitfallsApplicationsFrequently asked🔒 Read the full methodSourcesRelated methods
Cite this pageSpotted an issue on this page? Report or suggest a fix →
Home›Statistics›Robust Factor Analysis
Regression model

Robust Factor Analysis

Also known as: robust factor analysis, outlier-resistant factor analysis, MCD-based factor analysis, Robust Faktör Analizi

Robust Factor Analysis recovers the latent factor structure of multivariate continuous data while resisting the distorting pull of outliers. Introduced by Pison, Rousseeuw, Filzmoser and Croux (2003), it replaces the classical sample covariance with a robust estimator such as the Minimum Covariance Determinant (MCD) or an S-estimator before extracting factors.

ScholarGate
  1. Regression model
  2. v1
  3. 2 Sources
  4. PUBLISHED
Cite this page →
Tools & resources
Download slides
Learn & explore

Read the full method

Members only

Sign in with a free account to read this section.

Sign in

Method map

The neighbourhood of related methods — select a node to explore.

Robust Factor Analysis
Factor AnalysisInfluence DiagnosticsPrincipal Component Anal…Robust Covariance (MCD)Robust PCA

When to use it

Use Robust Factor Analysis with multivariate continuous data when you suspect outliers or non-normality and still want to uncover or describe a latent factor structure. It assumes continuous variables and is designed precisely for the situation where contamination or heavy tails would distort a classical solution. A reasonably large sample is needed — at least about 100 observations — because the robust covariance estimate must be reliable.

Strengths & limitations

Strengths
  • Extracts a factor structure that is not pulled toward outliers, unlike classical factor analysis.
  • Does not require multivariate normality, making it suitable for contaminated or heavy-tailed data.
  • Built on principled robust covariance estimators (MCD, S-estimators) with established breakdown properties.
Limitations
  • Requires a fairly large sample (about 100+) for the robust covariance to be trustworthy.
  • With fewer than 30 observations the robust covariance estimate becomes unreliable and classical PCA is preferable.
  • When the proportion of outliers is very high (above roughly 25%), even an MCD-based covariance can break down.

Frequently asked

How is this different from ordinary factor analysis?

The model is the same, but the covariance it factors is estimated robustly. Classical factor analysis uses the ordinary sample covariance, which a few outliers can distort; Robust Factor Analysis uses an MCD or S-estimator covariance that is computed from the bulk of the data, so the extracted factors are not driven by extreme cases.

What is the MCD estimator?

The Minimum Covariance Determinant fits the location and covariance to the subset of observations whose covariance has the smallest determinant — effectively the densest, most typical core of the data. Because outliers are excluded from that core, the resulting covariance, and the factors derived from it, resist contamination.

How large a sample do I need?

Plan for at least about 100 observations. Below roughly 30 the robust covariance estimate is not trustworthy, and classical PCA is the recommended fallback in that small-sample situation.

What if a large share of my data are outliers?

When the outlier ratio climbs above about 25%, even an MCD-based covariance can break down. In that regime robust PCA is the preferred alternative for recovering structure.

Sources

  1. Pison, G., Rousseeuw, P. J., Filzmoser, P., & Croux, C. (2003). Robust factor analysis. Journal of Multivariate Analysis, 84(1), 145-172. DOI: 10.1016/S0047-259X(02)00007-6 ↗
  2. Hubert, M., Rousseeuw, P. J., & Vanden Branden, K. (2005). ROBPCA: A new approach to robust principal component analysis. Technometrics, 47(1), 64-79. DOI: 10.1198/004017004000000563 ↗

How to cite this page

ScholarGate. (2026, June 1). Robust Factor Analysis. ScholarGate. https://scholargate.app/en/statistics/robust-factor-analysis

Related methods

Factor AnalysisInfluence DiagnosticsPrincipal Component AnalysisRobust Covariance (MCD)Robust PCA

Which method?

Set this method beside its closest kin and read them side by side — the library lays the books on the table; the choice is yours.

  • Factor AnalysisResearch Statistics↔ compare
  • Influence DiagnosticsStatistics↔ compare
  • Principal Component AnalysisMachine learning↔ compare
  • Robust Covariance (MCD)Statistics↔ compare
  • Robust PCAStatistics↔ compare
Compare side by side →

Similar methods

Robust Exploratory Factor AnalysisRobust Covariance (MCD)Robust Discriminant AnalysisRobust Mahalanobis DistanceRobust PCARobust Cluster AnalysisRobust Multiple Correspondence AnalysisRobust Confirmatory Factor Analysis

Related reference concepts

Factor AnalysisFactor AnalysisPrincipal Component AnalysisDimension ReductionMatrix Decompositions in StatisticsMultivariate Analysis of Variance

Spotted an issue on this page? Report or suggest a fix →

ScholarGate — Robust Factor Analysis (Robust Factor Analysis). Retrieved 2026-07-21 from https://scholargate.app/en/statistics/robust-factor-analysis · Dataset: https://doi.org/10.5281/zenodo.20539026
Quick facts
Originator
Pison, Rousseeuw, Filzmoser & Croux
Year
2003
Type
Robust latent-factor model
Estimator
MCD or S-estimator robust covariance, then factored
Outcome
latent factor structure
MinSample
100
Related methods
Factor AnalysisInfluence DiagnosticsPrincipal Component AnalysisRobust Covariance (MCD)Robust PCA
ScholarGate

A content-first reference library for research methods — what each one is, how it works, and where it comes from.

Open data (CC-BY)

Explore

  • Library
  • Search the library…
  • Browse by field
  • Fields
  • Journey
  • Compare
  • Which method?

Reference

  • Subjects
  • Atlas
  • Glossary
  • Methodology
  • Philosophy

Your tools

  • Bookshelf
  • Desk
  • Chat

Company

  • About
  • Pricing
  • Contact
  • Suggest a method

Entries are compiled from published sources for reference. Verifying the accuracy and suitability of any information for your own use remains your responsibility.

© 2026 ScholarGate · A research-method reference library
  • Privacy
  • Cookies
  • Terms
  • Delete account