calcvcox: Cross-validation for the penalized finite-mixture Cox PH...
In scrcss/pfmc_pkg: IRLS-EM algorithm for the penalized finite-mixture Cox PH model

Description Usage Arguments Value Author(s) References See Also Examples

Cross-validation for the choice of tuning parameters in the penalized finite-mixture Cox regression method

1
2
3

calcvcox(Time, Delta, X, ncore = 2, K = 2, nfolds = 10, 
         nopenaltyresult = NULL, alpha = 1, scad = FALSE, 
		 adpcoef = NULL, nlambda = 100, seed = 1)

`Time`	observed time
`Delta`	survival status
`X`	a data matrix of explanatory variables, where each colomn correponds to one predictor and each row indicates one sample.
`ncore`	the number of CPU cores occupied for parallele implementation
`K`	number of components in the finite-mixture Cox model
`nfolds`	number of folds split in CV partitioning
`nopenaltyresult`	fitting of the full finite-mixture Cox PH model, which is a list with `U`, a matrix of the posterior probabilities of each sample belonging to each component, and `pi`, the estimate for the mixing probability
`alpha`	the elastic net mixing parameter
`scad`	is the SCAD penalty applied?
`adpcoef`	the adaptive weight of adaptive LASSO method
`nlambda`	the number of `lambda` values
`seed`	random seeding for CV split

a data matrix, where the first row lambda.min presents the value of tuning parameter that gives the minimum CV error in each component of the finite-mixutre Cox PH model, and the second row lambda.1se provides the largest value of tuning parameter such that CV error is within 1 standard error of the minimum in each component.

Shijie Quan, Shun He

Subtype classification and heterogeneous prognosis model construction in precision medicine. Na You, Shun He, Xueqin Wang, Junxian Zhu and Heping Zhang

pmixcox