Partial Least Squares Regression for numerical variables.
PLSR(Y, X, S = 2, InitTransform = 5, grouping = NULL,
centerY = TRUE, scaleY = TRUE, tolerance = 5e-06,
maxiter = 100, show = FALSE, Validation = NULL, nB = 500)An object of class plsr with fiends
PLSR
The X matrix
The Y matrix
Is the Y matrix centered
Is the Y matrix scaled
Initial transformation of the Y matrix
Transformed X matrix
Transformed Y matrix
Intercept of the model
Scores for the individals from the X matrix
Weigths for the X set
Loadings for the X set
Scores for the individals from the Y matrix
Weigths for the Y set
Loadings for the Y set
Final Regression Parameters
Expected values of Y
R-squared
Relation of the X variables with its structure
Relation of the Y variables with its structure
Relation of the Y variables with the X components
Matrix of Dependent Variables
Matrix of Independent Variables
Dimension of the solution
Initial transformation of the independent variables.
Fator when the init transformation is the standardization with the within groups deviation.
Should the dependent variables be centered?
Should the dependent variables be standadized?
Tolerance for the algorithm
Maximum number of iterations
Show the progress of the algorithm?
Validation (None, Cross, Bootstrap)
number of samples for the bottstrap validation
Jose Luis Vicente Villardon
Partial Least Squares Regression for numerical variables.
H. Abdi, Partial least squares regression and projection on latent structure regression (PLS regression), WIREs Comput. Stat. 2 (2010), pp. 97-106.
Biplot.PLSR
X=as.matrix(wine[,4:21])
y=as.numeric(wine[,2])-1
mifit=PLSR(y,X, Validation="None")
Run the code above in your browser using DataLab