simple.ridge {ElemStatLearn} | R Documentation |
Function to do simple ridge regression.
simple.ridge(x, y, lambda = 1, df, ...)
x |
predictor matrix, NA's not allowed. Should not include a column of 1's. |
y |
response matrix, NA's not allowed |
lambda |
vector of ridge coefficients |
df |
vector of target degrees of freedom. lambda is calculated from these. |
... |
Presently not used |
Principal advantage with this compared to other implementations of ridge regression in R is that we can do the calculations for multiple degrees of freedom at once. Ridging is not applied to the intercept.
List with components:
beta0 |
intercept |
beta |
matrix of coefficients. One column for each lambda |
lambda |
lambda used |
df |
df used |
Kjetil B. Halvorsen
# See examples for prostate.