UCL  IRIS
Institutional Research Information Service
UCL Logo
Please report any queries concerning the funding data grouped in the sections named "Externally Awarded" or "Internally Disbursed" (shown on the profile page) to your Research Finance Administrator. Your can find your Research Finance Administrator at https://www.ucl.ac.uk/finance/research/rs-contacts.php by entering your department
Please report any queries concerning the student data shown on the profile page to:

Email: portico-services@ucl.ac.uk

Help Desk: http://www.ucl.ac.uk/ras/portico/helpdesk
Publication Detail
Variable selection for regression analysis: An up-date on an old problem
  • Publication Type:
    Journal article
  • Publication Sub Type:
    Journal Article
  • Authors:
    Davies AMC, Fearn T
  • Publication date:
    01/12/2012
  • Pagination:
    21, 23
  • Journal:
    Spectroscopy Europe
  • Volume:
    24
  • Issue:
    6
  • Status:
    Published
  • Print ISSN:
    0966-0941
Abstract
Powered partial least squares (PPLS) is a modification of the NIPALS algorithm, which was one of the earliest of the many algorithms that now exist for PLS. The modification concerns the computation of the loading weights, which determine the contributions of the original variables to each factor. In the standard algorithm these loading weights are proportional to the covariances between the dependent variable y and the predictors. PPLS gives more weight to the correlations, with an extra tuning parameter to control this weight. The influence of a variable on the prediction result depends on the size of the coefficient computed by PLS. If the coefficient is very small then it will have very little effect on the result except that every variable will add a small noise component. Thus predictions produced by PLS calibrations must have more than the minimum of added noise. If it was possible to include variables in PLS only when they made a useful contribution to the precision of the predicted result then some of this noise could be excluded.
Publication data is maintained in RPS. Visit https://rps.ucl.ac.uk
 More search options
UCL Researchers
Author
Dept of Statistical Science
University College London - Gower Street - London - WC1E 6BT Tel:+44 (0)20 7679 2000

© UCL 1999–2011

Search by