| coef.mvr {pls} | R Documentation |
Functions to extract the regression coefficients or the model matrix
from mvr objects.
## S3 method for class 'mvr':
coef(object, comps = object$ncomp, intercept = FALSE,
cumulative = TRUE, ...)
## S3 method for class 'mvr':
model.matrix(object, ...)
object |
an mvr object. The fitted model. |
comps |
vector of positive integers. The components to include in the coefficients. Defaults to the number of components used for fitting the model. |
intercept |
logical. Whether coefficients for the intercept should
be included. Ignored if cumulative = FALSE. Defaults to
FALSE. |
cumulative |
logical. Whether cumulative (the default) or individual coefficients for each component should be returned. See below. |
... |
other arguments sent to underlying functions. Currently
only used for model.matrix.mvr. |
coef.mvr is used to extract the regression coefficients of a
model, i.e. the B in y = XB. An array of dimension
c(nxvar, nyvar, length(comps)) is returned.
If cumulative = TRUE, coef()[,,comps[i]] are
the coefficients for models with comps[i] components, for
i = 1, ..., length(comps). Also, if intercept = TRUE,
the first dimension is nxvar + 1, with the intercept
coefficients as the first row.
If cumulative = FALSE, however, coef()[,,comps[i]] are
the coefficients for a model with only the component comps[i],
i.e. the contribution of the component comps[i] on the
regression coefficients.
model.matrix.mvr returns the (possibly coded) matrix used as
X in the fitting.
coef.mvr returns an array of regression coefficients.
model.matrix.mvr returns the X matrix.
Ron Wehrens and Bjørn-Helge Mevik
data(NIR)
mod <- pcr(y ~ X, data = NIR[NIR$train,], ncomp = 5)
B <- coef(mod, comps = 3, intercept = TRUE)
## A manual predict method:
stopifnot(drop(B[1,,] + NIR$X[!NIR$train,] %*% B[-1,,]) ==
drop(predict(mod, comps = 3, newdata = NIR[!NIR$train,])))