Residuals From a Fitted Arbor Object
USAGE:
residuals.arbor(object, type)
REQUIRED ARGUMENTS:
- object
-
fitted model object of class
arbor. This is assumed to be the result
of some function that produces an object with the same named
components as that returned by the
arbor function.
OPTIONAL ARGUMENTS:
- type
-
Indicates method of residual for a classification tree (no distinction
is made for methods
"anova",
"poisson", or
"exp"). Allows the
following types of residuals:
"usual" (misclassification errors),
"pearson" (sum of the observed minus fitted squared over fitted), and
"deviance" (minus twice logarithm of fitted).
VALUE:
vector of residuals of type
type from a fitted
arbor object.
EXAMPLES:
fit <- arbor(skips ~ Opening + Solder + Mask + PadType + Panel,
data=solder, method='anova')
plot(predict(fit),residuals(fit))