Compute failure probability estimates

DESCRIPTION:

Computes Turnbull's generalization (EM) of the Kaplan-Meier estimate.

USAGE:

kaplanMeier.fit(y, censor.codes, case.weights, maxit=500, tol=0.001, maxmsd=10) 

REQUIRED ARGUMENTS:

y
matrix of failure or censoring intervals. Except for interval censored data, y has two equal columns. For interval censoring, the first column contains the start times, and the second column contains the ending times for the interval.
censor.codes
vector of censoring codes. These are: 1 exact failure 2 right censored 3 left censored 4 interval censored
case.weights
vector of case weights.

OPTIONAL ARGUMENTS:

maxit
maximum number of iterations.
tol
When the absolute change in all failure probabilities is less than tol, convergence is assumed.
maxmsd
the maximum number of intervals for which information matrix based standard errors will be computed.

VALUE:

A matrix containing the survival probabilities.

SEE ALSO:

EXAMPLES:

This routine is not normally called by users.