residuals.survreg {survival}  R Documentation 
This is a method for the function residuals
for objects
inheriting from class survreg
.
residuals.survreg(object, type=c("response", "deviance","dfbeta","dfbetas", "working","ldcase","ldresp","ldshape", "matrix"), rsigma=TRUE, collapse=FALSE, weighted=FALSE, ...)
object 
an object inheriting from class survreg .

type 
type of residuals, with choices of "response" , "deviance" ,
"dfbeta" , "dfbetas" , "working" , "ldcase" , "lsresp" ,
"ldshape" , and "matrix" . See the LaTeX documentation
(survival/doc/survival.ps.gz ) for more
detail.

rsigma 
include the scale parameters in the variance matrix, when doing computations. (I can think of no good reason not to). 
collapse 
optional vector of subject groups. If given, this must be of the same length as the residuals, and causes the result to be per group residuals. 
weighted 
give weighted residuals? Normally residuals are unweighted. 
... 
other unused arguments 
A vector or matrix of residuals is returned.
Response residuals are on the scale of the original data,
working residuals are on the scale of the linear predictor,
and deviance residuals are on loglikelihood scale.
The dfbeta residuals are a matrix, where the ith row gives the
approximate change in the coefficients due to the addition of subject i.
The dfbetas matrix contains the dfbeta residuals, with each column
scaled by the standard deviation of that coefficient.
The matrix type produces a matrix based on derivatives of the loglikelihood
function. Let L be the loglikelihood, p be the linear predictor X %*% coef,
and s be log(σ). Then the 6 columns of the matrix are
L, dL/dp,ddL/(dp dp),
dL/ds, ddL/(ds ds) and
ddL/(dp ds). Diagnostics based on these quantities
are discussed in an article by Escobar and Meeker.
The main ones are the likelihood displacement residuals for perturbation
of a case weight (ldcase
), the response value (ldresp
), and the shape
.
Escobar, L. A. and Meeker, W. Q. (1992). Assessing influence in regression analysis with censored data. Biometrics 48, 507528.
fit < survreg(Surv(time,status) ~x, aml) rr < residuals(fit, type='matrix')