Jove
Visualize
Contact Us

Related Experiment Videos

Confidence intervals for the conditional probability of misallocation in discriminant analysis.

G J McLachlan

    Biometrics
    |March 1, 1975
    PubMed
    Summary

    This study introduces a new, accurate method for calculating confidence intervals for misallocation probability in Anderson

    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

    Clustering.

    Methods in molecular biology (Clifton, N.J.)·2016
    Same author

    A benchmark for evaluation of algorithms for identification of cellular correlates of clinical outcomes.

    Cytometry. Part A : the journal of the International Society for Analytical Cytology·2015
    Same author

    Clustering of gene expression data via normal mixture models.

    Methods in molecular biology (Clifton, N.J.)·2013
    Same author

    Testing for group structure in high-dimensional data.

    Journal of biopharmaceutical statistics·2011
    Same author

    Application of gene shaving and mixture models to cluster microarray gene expression data.

    Cancer informatics·2009
    Same author

    A mixture model with random-effects components for clustering correlated gene-expression profiles.

    Bioinformatics (Oxford, England)·2006
    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

    Area of Science:

    • Statistics
    • Machine Learning
    • Classification Analysis

    Background:

    • Existing methods for confidence intervals of misallocation probability are often inaccurate in practice.
    • Anderson's classification statistics (W) are widely used but require reliable confidence intervals.
    • Accurate confidence intervals are crucial for evaluating classification model performance.

    Purpose of the Study:

    • To develop a more accurate and practical method for constructing confidence intervals for the conditional probability of misallocation.
    • To address the limitations of current confidence interval computation methods for Anderson's W statistic.

    Main Methods:

    • A novel approach for computing confidence intervals is presented.
    • The method utilizes the initial samples used for Anderson's W statistic.
    • Focus on ease of computation and achieving desired confidence levels.

    Main Results:

    • The proposed method provides confidence intervals with improved accuracy.
    • The new intervals are easily computed from the base data.
    • Achieves near-desired confidence levels, enhancing practical utility.

    Conclusions:

    • The new method offers a significant improvement over existing techniques for confidence intervals of misallocation probability.
    • This facilitates more reliable assessment of classification performance using Anderson's W.
    • The approach is practical for researchers and practitioners.

    Related Experiment Videos