Probabilistic predictions for partial least squares using bootstrap

Abstract

Modeling the uncertainty in partial least squares (PLS) is made difficult because of the nonlinear effect of the observed data on the latent space that the method finds.We present an approach, based on bootstrapping, that automatically accounts for these nonlinearities in the parameter uncertainty, allowing us to equally well represent confidence intervals for points lying close to or far away from the latent space.To show the opportunities of this approach, we develop applications in determining the Design Space for industrial processes and model the uncertainty of spectroscopy data. Our results show the benefits of our method for accounting for uncertainty far from the latent space for the purposes of Design Space identification, and match the performance of well established methods for spectroscopy data.

An Optimization Approach Coupling Preprocessing with Model Regression for Enhanced Chemometrics

“…As previously mentioned, the use of RMSEP is commonly utilized in this regard. , Alternative/complementary metrics for accuracy, such as root-mean-square error of calibration, or cross validation and/or bias are conceivable. These measures are sufficient for understanding the average errors in a data set; however they should not be interpreted as an estimate for the uncertainty of a specific prediction at a new input . Moreover, use of these measures ignores the fact that errors can come from different distributions that have the same mean error but different shapes …”

“…These measures are sufficient for understanding the average errors in a data set; however they should not be interpreted as an estimate for the uncertainty of a specific prediction at a new input. 39 Moreover, use of these measures ignores the fact that errors can come from different distributions that have the same mean error but different shapes. 40 Precision can be measured with repeatability, i.e., standard deviation of repeated measurements of the same sample, where the sample is not moved during spectral acquisition.…”

Product Design for Batch Processes Based on Iterative Learning-Latent Variable Model Inversion (IL-LVMI)

“…However, creating an accurate process model can be intimidating and impractical, as it relies heavily on extensive process knowledge and requires frequent model adaptation to changes in process dynamics. − As an alternative, data-driven approaches have been developed to speed up the design process . These methods typically rely on large amounts of historical or experimental data and can generally be classified into two categories: deterministic − and probabilistic methods, − the former does not consider model uncertainty whereas the latter does …”