May 2019
Beginner to intermediate
266 pages
5h 57m
English
Partial Least Squares Regression (PLSR) creates a linear model of the data in a transformed projection of a problem space. Like PCR, PLSR is appropriate for data with highly correlated predictors, shown as follows:
> plsFit<-plsr(Employed ~., data = longley, validation = "CV")> #summarise the fit model> summary(plsFit)Data: X dimension: 16 6 Y dimension: 16 1Fit method: kernelplsNumber of components considered: 6VALIDATION: RMSEPCross-validated using 10 random segments. (Intercept) 1 comps 2 comps 3 comps 4 comps 5 comps 6 compsCV 3.627 1.415 1.063 0.5404 0.6336 0.4696 0.4197adjCV 3.627 1.397 1.052 0.5324 0.6168 0.4644 0.4076TRAINING: % variance explained 1 comps 2 comps 3 comps 4 comps 5 comps 6 compsX 63.88 ...
Read now
Unlock full access