Hi! 

I am trying to fit a Cox model with imputed data. However, the calibration
function does not seem to work after applying fit.mult.impute. 
Secondly, which is the best method to validate the model? 

This is the code... 

Could anyone help, please? 

aga_imp =aregImpute
(~her2_cox+ecog_cox+bone_cox+nmet3_cox+ascitis_cox+Grade+nlr_cox,data
=aga_NA, x=T, nk=0, n.impute =5) 
ddist=datadist(aga_sinNA) 
options(datadist="ddist") 
suv<- with(aga_NA, Surv(time,event)) 
cox <- fit.mult.impute(suv ~
her2_cox+ecog_cox+bone_cox+nmet3_cox+ascitis_cox+Grade+nlr_cox, cph,
aga_imp, data=aga_NA, n.impute=5, pr=F, surv=T, time.inc=365 ) 
cal<-calibrate(cox, cmethod='KM', method='boot',u=365,B=10) 


Unfortunately I obtain this: 
Error in predab.resample(fit, method = method, fit = coxfit, measure =
distance,  : 
  must have specified x=T and y=T on original fit 




--
View this message in context: 
http://r.789695.n4.nabble.com/Calibrate-after-fit-mult-impute-for-a-Cox-model-rms-package-tp4725353.html
Sent from the datatable-help mailing list archive at Nabble.com.
_______________________________________________
datatable-help mailing list
[email protected]
https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/datatable-help

Reply via email to