Parametric and Nonparametric Approaches of Reid Vapor Pressure Prediction for Gasoline Containing Oxygenates: A Comparative Analysis Using Partial Least Squares, Nonlinear, and LOWESS Regression Modelling Strategies with Physical Properties

Figure 4

Normality assumption plots for PLSR prediction performance: (a) histogram plot for residuals and (b) probability plot of residual for semistudentized residuals (), where is the standardized residuals, is the th ordered residual, and is the inverse normal distribution of ordered residuals, and (c) two-dimensional plot of PCA latent variables LV1 and LV2 loading with PLSR model for response and explanatory variables.

(a)

(b)

(c)