Access

You are not currently logged in.

Access your personal account or get JSTOR access through your library or other institution:

login

Log in to your personal account or through your institution.

An MM Algorithm for Multicategory Vertex Discriminant Analysis

Kenneth Lange and Tong Tong Wu
Journal of Computational and Graphical Statistics
Vol. 17, No. 3 (Sep., 2008), pp. 527-544
Stable URL: http://www.jstor.org/stable/27594324
Page Count: 18
  • Download ($14.00)
  • Cite this Item
An MM Algorithm for Multicategory Vertex Discriminant Analysis
Preview not available

Abstract

This article introduces a new method of supervised learning based on linear discrimination among the vertices of a regular simplex in Euclidean space. Each vertex represents a different category. Discrimination is phrased as a regression problem involving ∈-insensitive residuals and a quadratic penalty on the coefficients of the linear predictors. The objective function can by minimized by a primal MM (majorization—minimization) algorithm that (a) relies on quadratic majorization and iteratively reweighted least squares, (b) is simpler to program than algorithms that pass to the dual of the original optimization problem, and (c) can be accelerated by step doubling. Limited comparisons on real and simulated data suggest that the MM algorithm is competitive in statistical accuracy and computational speed with the best currently available algorithms for discriminant analysis.

Page Thumbnails

  • Thumbnail: Page 
527
    527
  • Thumbnail: Page 
528
    528
  • Thumbnail: Page 
529
    529
  • Thumbnail: Page 
530
    530
  • Thumbnail: Page 
531
    531
  • Thumbnail: Page 
532
    532
  • Thumbnail: Page 
533
    533
  • Thumbnail: Page 
534
    534
  • Thumbnail: Page 
535
    535
  • Thumbnail: Page 
536
    536
  • Thumbnail: Page 
537
    537
  • Thumbnail: Page 
538
    538
  • Thumbnail: Page 
539
    539
  • Thumbnail: Page 
540
    540
  • Thumbnail: Page 
541
    541
  • Thumbnail: Page 
542
    542
  • Thumbnail: Page 
543
    543
  • Thumbnail: Page 
544
    544