Sfn-class: A Class containing the Survival Function (for a single arm)...

In scientific-computing-solutions/eventPrediction: Event Prediction in Clinical Trials with Time-to-Event Outcomes

Description

A Class containing the Survival Function (for a single arm) used in the integral to calculate event times in predict from parameters. It should not be created by the user but is created for the user (see sfns slot of AnalysisResult)

Slots

sfn

A function sfn(x) is the function to be included in the events.integrate procedure when using predict from parameters. Specifically sfn(x) = 1 - P(had event [i.e. not dropout] by time x) and when no drop outs this is exactly the survival function. If using dropouts this is not quite the survival function However, the output of LatexSurvivalFn(object) is the survival function.

SurvivalFunction

The actual survival function (if drop outs/finite followup are used then this will not equal sfn). If finite follow up is used then S(x) = 0 for x > followup

pdf

The pdf function associated with the survival function. If finite follow up is used then pdf(x) should = 0 for x > followup

nullf

Logical, TRUE if the object represents NULL (i.e. a survival function for a second arm in a single arm study)

lambda

The rate parameters for the arm of trial. In a trial with lag, this is the rate parameters for time > T

lambdaot

If a lag was used then the rate parameters for time < T otherwise NA

shape

The Weibull shape parameter

LagT

The lagtime for the survival function (0 for no lag)

followup

The follow up time for each subject (Inf for studies with no fixed followup)

dropout.shape

The Weibull shape parameter for the drop out hazard function

dropout.lambda

The rate parameter for the drop out hazard function = 0 if no dropout

scientific-computing-solutions/eventPrediction documentation built on May 29, 2019, 3:44 p.m.