es {geecure} | R Documentation |
Expectation-Solution (ES) algorithm
Description
ES algorithm is an estension of the EM algorithm where the M-step of the EM algorithm is replaced by a step requiring the solution of a series of generalised estimating equations. We use the ES algorithm for the analysis of survival cure data with potential correlation.
Usage
es(Time, Status, X, Z, id, model, corstr, stdz, esmax, eps)
Arguments
Time |
right censored data which is the follow up time. |
Status |
the censoring indicator, normally 0 = event of interest happens, and 0 = censoring. |
X |
a matrix of covariates corresponding to the latency part. |
Z |
a matrix of covariates corresponding to the incidence part. |
id |
a vector which identifies the clusters. The length of |
model |
specifies your model, it can be |
corstr |
a character string specifying the correlation structure. The following are permitted: |
stdz |
If it is TRUE, all the covariates in the |
esmax |
specifies the maximum iteration number. If the convergence criterion is not met, the ES iteration will be stopped after |
eps |
tolerance for convergence. The default is |