Iterative Predictor Weighting PLS (IPW): A technique for the elimination of useless predictors in regression problems

  1. Forina, M. 1
  2. Casolino, C. 1
  3. Pizarro Millan, C. 2
  1. 1 Dipartimento di Chimica e Tecnologie Farmaceutiche ed Alimentari, Università di Genova, Via Brigata Salerno (Ponte), I-16147 Genova, Italy
  2. 2 Universidad de La Rioja
    info

    Universidad de La Rioja

    Logroño, España

    ROR https://ror.org/0553yr311

  3. 3 Dipartimento di Chimica e Tecnologie Farmaceutiche ed Alimentari, Via Brigata Salerno (Ponte), I-16147 Genova, Italy
Mostrar afiliaciones +
Revista:
Journal of Chemometrics

ISSN: 0886-9383

Año de publicación: 1999

Volumen: 13

Número: 2

Páginas: 165-184

Tipo: Artículo

SCOPUS: 2-s2.0-0001286453 WoS: WOS:000080425000008 DIALNET GOOGLE SCHOLAR

Otras publicaciones en: Journal of Chemometrics

Repositorio institucional: lock_openAcceso abierto Editor

Resumen

A new method for the elimination of useless predictors in multivariate regression problems is proposed. The method is based on the cyclic repetition of PLS regression. In each cycle the predictor importance (product of the absolute value of the regression coefficient and the standard deviation of the predictor) is computed, and in the next cycle the predictors are multiplied by their importance. The algorithm converges after 10-20 cycles. A reduced number of relevant predictors is retained in the final model, whose predictive ability is acceptable, frequently better than that of the model built with all the predictors. Results obtained on many real and simulated data are presented, and compared with those obtained from other techniques. Copyright © 1999 John Wiley & Sons, Ltd.