A robust functional partial least squares for scalar?on?multiple?function regression
Abstract
The scalar-on-function regression model has become a popular analysis tool to explore the relationship between scalar response and multiple functional predictors. Most of existing approaches estimate this are based on least-squares estimator, which can be seriously affected by outliers in empirical datasets. When present data, it is known that least-squares-based estimates may not reliable. This paper proposes robust partial least squares method, allowing coefficients scalar-on-multiple-function model. In our components computed via M-regression. predictive performance proposed method evaluated using several Monte Carlo experiments two chemometric datasets: glucose concentration spectrometric data sugar process data. results produced compared favorably with some classical or multivariate principal component methods.