Multicomponent determination of pesticides in vegetables by gas chromatography with mass spectrometric detection and multivariate calibration

Prediction of Relative Density, Distillation Temperatures, Flash Point, and Cetane Number of S500 Diesel Oil Using Multivariate Calibration of Gas Chromatographic Profiles

“…However, it is worth pointing out that randomized measurements of the variables ( X , Y , etc.) from an unsuitable preprocessing can generate unclear results when building a predictive model …”

“…from an unsuitable preprocessing can generate unclear results when building a predictive model. 42 The first step to the development of a predictive model begins from the selection of an appropriate number of latent variables (LVs). In this sense, to meet the specific needs of a more effective predictive model, it is necessary to create distinct calibration models, whose LVs change from one model to another, so that predictive ability for each model is given by the statistical parameters SEC (Standard Error of Calibration) and SEV (Standard Error Validation).…”