is it possible to apply "sure " R package to ordinal models "svyolr"?

[fab2001]

Dear prof. Greenwell,

I am trying to use “sure” package to make diagnostics on a logit ordinal model generated with “svyolr” function (“survey” R package”), in particular i am trying to use “resids” function, but i get the error “Error in eval(predvars, data, env) : object "MR" not found”, where “MR” is the response variable (ordered factor) of the svyolr model.

I tryed to assign the class “polr” to the svyolr model (that is based on polr function) but the error persists.

I apologize if the question is trivial but I can’t get rid of the error.

Yours sincerely,

Please find below an example.

Fabio Maria Colombo

---

library(sure)
library(polr)
library(survey)

#Simulated dataset
RM=ordered(c(0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 3, 3, 3, 3, 1, 1, 1, 1, 1, 2, 2, 2, 2, 3, 3, 3, 3,
3, 3, 3, 3, 4, 4, 4, 3, 3, 4, 4, 4, 4))
data=data.frame(age=rnorm(mean=7,sd=2,49),
weight=rnorm(mean=10,sd=3,49),
nose.length=rnorm(49,mean=3,sd=1.5),
head.length=rnorm(49,mean=8,sd=2),
w=abs(1+trunc(rnorm(mean=2, sd=2,49))))
#survey design
svydesign<-svydesign(id=~1,data=data,weights=~w)

#svyolr model
model=svyolr(RM~nose.length+head.length,
design=svydesign
)
#resids(sure) function
resids(model)

Error in resids.default(model) :

model should be of class "clm", "glm", "lrm", "orm", "polr", "vgam", or "vglm".

class(model)='polr' #assigns the class "polr" to model
resids(model)
#Error in eval(predvars, data, env) : object "nose.length" not found
####################################################################