Rsqr.regpath: R^2 regularization path for ridge RDA

In pr2: Predictive R-squared

Description

R^2 regularization path for ridge RDA

Usage

  Rsqr.regpath(Y, X, form, trans = "hellinger",
    lambdas = exp(-seq(0, ncol(X) + 1, length.out = 30)),
    ...)

Arguments

Y	Response matrix
X	Design matrix
form	Optional one-sided formula for specifying Design matrix
trans	Transformation of the response matrix
lambdas	Vector of regularization parameters defining the path
...	Additional parameters

Value

A data frame with one row per lambda and columns:

lambdas	lambdas
df	Effective degrees of freedom (i.e. trace of the hat matrix)
rel.eff.n.params	Relative effective number of parameters
Rsquare	Unadjusted R^2
Rsquare.adj	Adjusted R^2
Rsquare.pred	Doubly adjusted (i.e. prediction) R^2

pr2 documentation built on May 2, 2019, 6:12 p.m.