Displaying 1 result from an estimated 1 matches for "bootcv".
Did you mean:
bootcov
2013 Jul 06
1
problem with BootCV for coxph in pec after feature selection with glmnet (lasso)
...t (glmnet.cox) ----> 354 df, p=1 n= 368, number of events= 288 (354 df
= 354 variables + time and status => 356 variables)
Here is the pec function and the error:
pec.f <- as.formula(Hist(time,status) ~ 1)
brierGlmnet <- pec(list(glmnet.cox), data = reformat_Dataset,
splitMethod="BootCV", B=50, formula = pec.f)
> Error in predictSurvProb.coxph(object = structure(list(coefficients =
structure(c(-4.27787223119601, :
Prediction matrix has wrong dimensions:
368 rows and 318 columns.
But requested are predicted probabilities for
118 subjects (rows) in newdata...