Explorative methods in linear models

Agnar Høskuldsson

    Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

    Abstract

    The author has developed the H-method of mathematical modeling that builds up the model by parts, where each part is optimized with respect to prediction. Besides providing with better predictions than traditional methods, these methods provide with graphic procedures for analyzing different features in data. These graphic methods extend the well-known methods and results of Principal Component Analysis to any linear model. Here the graphic procedures are applied to linear regression and Ridge Regression.
    Original languageEnglish
    Title of host publicationSymposium i Anvendt Statistik
    EditorsAnders Holm
    Publication date2004
    Publication statusPublished - 2004
    EventSymposium i Anvendt Statistik 2004 - København, Denmark
    Duration: 5 Nov 2004 → …

    Conference

    ConferenceSymposium i Anvendt Statistik 2004
    CountryDenmark
    CityKøbenhavn
    Period05/11/2004 → …

    Cite this

    Høskuldsson, A. (2004). Explorative methods in linear models. In A. H. (Ed.), Symposium i Anvendt Statistik http://www.ipl.dtu.dk/publikation/7912/dk/