Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Robust principal components: a generalized variance perspective.

Rand R Wilcox1

  • 1Department of Psychology, University of Southern California, Los Angeles, California 90089-1061, USA. rwilcox@usc.edu

Behavior Research Methods
|April 17, 2008
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

From significant to meaningful: ATOMizing the study of sex differences and similarities.

Frontiers in neuroendocrinology·2026
Same author

Influence of Body Configuration on Kinetics and Multijoint Control Strategies Sprinters Use During the First Step Out of Blocks.

Journal of applied biomechanics·2026
Same author

Associations of sleep behaviors with white matter hyperintensity volume in middle-aged to older adults.

Alzheimer's & dementia : the journal of the Alzheimer's Association·2026
Same author

Public Health.

Alzheimer's & dementia : the journal of the Alzheimer's Association·2025
Same author

From significant to meaningful: ATOMizing the study of sex differences and similarities.

Frontiers in neuroendocrinology·2025
Same author

Chronic inflammation mediates the relationship between physical activity and telomere length.

GeroScience·2025
Same journal

Planned missingness in intensive longitudinal studies: Extensions and comparisons of multiform designs.

Behavior research methods·2026
Same journal

A validity-guided workflow for robust large language model research in psychology.

Behavior research methods·2026
Same journal

Are 7-point Likert scales preferable to 5-point scales in language research?

Behavior research methods·2026
Same journal

Generative psychometrics via AI-GENIE: Automatic item generation and validation with network-integrated evaluation.

Behavior research methods·2026
Same journal

Exploring psychological tradeoffs: Developing and demonstrating an R Shiny app for Pareto optimization.

Behavior research methods·2026
Same journal

The performance of Bayesian fit measures in detecting misspecified multilevel structural equation modeling.

Behavior research methods·2026
See all related articles

This study compares robust principal component analysis methods, evaluating their ability to maximize projected data variance. Several robust methods, including two novel approaches, outperform existing techniques, especially for non-elliptically symmetric data.

Area of Science:

  • Statistics
  • Data Analysis

Background:

  • Principal Component Analysis (PCA) is a dimensionality reduction technique.
  • Robust methods are needed to handle outliers and non-elliptical data distributions in PCA.

Purpose of the Study:

  • To compare existing and novel robust principal component analysis (RPCA) methods.
  • To evaluate methods based on maximizing robust generalized variance of projected data.

Main Methods:

  • Comparison of RPCA methods including Maronna (2005), spherical method (Locantore et al., 1999), Hubert et al. (2005), and two new methods.
  • One new method uses multivariate outlier detection; the other is a novel projection pursuit technique.
  • Evaluation criterion: maximization of robust generalized variance, not just marginal scatter.

Related Experiment Videos

Main Results:

  • The Hubert et al. (2005) method, the spherical method, and one of the new methods demonstrated superior performance.
  • These methods dominated the Maronna (2005) method under the chosen criterion.
  • Comparisons included non-elliptically symmetric distributions.

Conclusions:

  • Novel robust principal component analysis methods show significant promise.
  • The choice of criterion (robust generalized variance) is crucial for evaluating RPCA methods.
  • Existing robust methods like Hubert et al. and the spherical method remain competitive.