Get residuals of various types
# S3 method for AccurateGLM
residuals(
object,
x = NULL,
y = NULL,
offset = NULL,
weights = NULL,
type = c("working", "pearson", "deviance"),
s = NULL,
...
)A numeric vector representing calculated residuals.
A model object obtained from aglm() or cv.aglm().
A design matrix.
If not given, x for fitting is used.
A response variable.
If not given, y for fitting is used.
An offset values.
If not given, offset for fitting is used.
Sample weights.
If not given, weights for fitting is used.
A string representing type of deviance:
"working" get working residual
r^W_i = (y_i - _i) ( )_=_i,
where \(y_i\) is a response value, \(\mu\) is GLM mean, and \(\eta=g^{-1}(\mu)\) with the link function \(g\).
"pearson" get Pearson residuals
r^P_i = y_i - _iV(_i),
where \(V\) is the variance function.
"deviance" get deviance residuals
r^D_i = sign(y_i - _i) d_i,
where \(d_i\) is the contribution to deviance.
A numeric value specifying \(\lambda\) at which residuals are calculated.
Other arguments are currently not used and just discarded.
Kenji Kondo