Implementing PLS for distance-based regression: computational issues

DOI10.1007/S00180-007-0035-2MaRDI QIDQ964602zbMATH OpenOpenAlexFDO

Authors Eva Boj, Aurea Grané, Josep Fortiana, M. Mercè Claramunt

Publication date 22 April 2010

Published in Computational Statistics (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1007/s00180-007-0035-2

zbMATH Keywords

moderately large samplemultidimensional scalingpartial least squares

Mathematics Subject Classification ID

Computational methods for problems pertaining to statistics (62-08) Multivariate analysis (62H99) Linear inference, regression (62J99) Applications of statistics to actuarial sciences and financial mathematics (62P05)

Review text

An algorithm is described for distance-based partial least squares regression. The proposed technique allows to calculate the principal coordinates avoiding explicit Euclidian configuration calculations which, for large datasets, are computationally expensive. Details of computer implementation are discussed for moderately large samples (of size \(\approx 10^4\)) when the distances matrix can be several gigabyte-sized. An application is considered to the problem of rate-making in automobile insurance.

Recommendations

• DPLS and PPLS: two PLS algorithms for large data sets
• A Reformulation of the Partial Least Squares Regression Algorithm
• PLS for Big Data: a unified parallel algorithm for regularised group PLS
• scientific article; zbMATH DE number 1058079
• On the PLS algorithm for multiple regression (PLS1)

Cites work

• Adding a point to vector diagrams in multivariate analysis
• scientific article; zbMATH DE number 18876 (Why is no real title available?)
• Modern multidimensional scaling. Theory and applications.
• On the structure of partial least squares regression
• Principal component analysis.
• Some computational aspects of a distance—based model for prediction

Cited in

(6)

• Distance-based beta regression for prediction of mutual funds
• Global and local distance-based generalized linear models
• Selection of Predictors in Distance-Based Regression
• Distance-based approach in univariate longitudinal data analysis
• A kernel distance-based representative subset selection method
• Functional \(k\)-sample problem when data are density functions

This page was built for publication: Implementing PLS for distance-based regression: computational issues

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q964602)