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

Regularized linear discriminant analysis and its application in microarrays.

Yaqian Guo1, Trevor Hastie, Robert Tibshirani

  • 1Department of Statistics, Stanford University, Stanford, CA 94305, USA. yaqiang@stanford.edu

Biostatistics (Oxford, England)
|April 11, 2006
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

Cell Cycle Sensing Shapes Human T Cell Fate and Exhaustion Programs.

bioRxiv : the preprint server for biology·2026
Same author

Wavelet Decomposition-Based Genomic Analysis of the Human Electrocardiogram.

medRxiv : the preprint server for health sciences·2026
Same author

Quantifying Anterior Cruciate Ligament Injury Resilience: A Screening and Composite Score Framework.

Orthopaedic journal of sports medicine·2026
Same author

Structure-preserving multivariate hypothesis testing for mass spectrometry imaging and single-cell data.

Bioinformatics (Oxford, England)·2026
Same author

Aptamer-targeted hybrid nanoparticles based on human exosomes and LXR agonist-loaded liposomes for enhanced anti-AML therapy.

Nucleosides, nucleotides & nucleic acids·2026
Same author

Temporal and spatial composition of the tumor microenvironment predicts response to immune checkpoint inhibition in metastatic TNBC.

Nature cancer·2026

Shrunken Centroids Regularized Discriminant Analysis (SCRDA) offers a new approach for high-dimensional classification problems with limited samples. This method shows strong performance, outperforming existing techniques and aiding in feature selection for complex datasets.

Area of Science:

  • Statistics
  • Machine Learning
  • Bioinformatics

Background:

  • Classical discriminant analysis methods often struggle with high-dimensional data and limited sample sizes.
  • Existing algorithms like PAM (Partitioning Around Medoids) using Nearest Shrunken Centroids (NSC) have limitations in certain complex classification tasks.

Purpose of the Study:

  • To introduce a modified linear discriminant analysis method, Shrunken Centroids Regularized Discriminant Analysis (SCRDA).
  • To address classification challenges in high-dimension, low-sample size (HDLSS) scenarios, such as those encountered with microarray data.

Main Methods:

  • Generalizing the Nearest Shrunken Centroids (NSC) concept within classical discriminant analysis.
  • Developing a regularization technique tailored for discriminant analysis in HDLSS settings.

Related Experiment Videos

Main Results:

  • SCRDA demonstrates robust performance in multivariate classification tasks using both simulated and real-world data.
  • The method often outperforms the PAM algorithm and shows competitive results compared to support vector machine classifiers.
  • SCRDA proves effective for feature elimination and can be utilized as a gene selection technique.

Conclusions:

  • SCRDA is a powerful and versatile tool for classification in high-dimensional, low-sample size environments.
  • The method offers advantages in performance and feature selection capabilities over existing approaches.
  • An open-source R package ('rda') is available for implementing SCRDA.