GSTobj {AGSDest}R Documentation

Group sequential trial object (GSTobj)

Description

The GSTobj includes design and outcome of primary trial.

Usage

GSTobj(x, ...)
## S3 method for class 'GSTobj':
print(x, ...)
## S3 method for class 'GSTobj':
plot(x,main="GSD",print.pdf=FALSE, ...)
## S3 method for class 'GSTobj':
summary(object,ctype="b",ptype="b",etype="b",overwrite=FALSE,...)
## S3 method for class 'summary.GSTobj':
print(x, ...)

Arguments

x object of the class GSTobj
object object of the class GSTobj
main Title of the plots (default: "GSD")
print.pdf option; if TRUE a pdf file is created. Instead of setting print.pdf to TRUE, the user can specify a character string giving the name or the path of the file.
ctype confidence type: repeated "r", stage-wise ordering "so", both "b" or none "n" (default: "b")
ptype p-value type: repeated "r", stage-wise ordering "so", both "b" or none "n" (default: "b")
etype point estimate: maximum likelihood "ml", median unbiased "mu", both "b" or none "n" (default: "b")
overwrite option; if TRUE all old values are deleted and new values are calculated (default: FALSE)
... additional arguments.

Details

A GSTobj object is designed.

The function summary returns an object of class GSTobj.

ctype defines the type of confidence interval that is calculated.
"r" Repeated confidence bound for a classical GSD
"so" Confidence bound for a classical GSD based on the stage-wise ordering
"b" both: repeated confidence bound and confidence bound based on the stage-wise for a classical GSD
"n" no confidence bound is calculated

The calculated confidence bounds are saved as:
cb.r repeated confidence bound
cb.so confidence bound based on the stage-wise ordering

ptype defines the type of p-value that is calculated.
"r" Repeated p-value for a classical GSD
"so" Stage-wise adjusted p-value for a classical GSD
"b" both: repeated and stage-wise adjusted p-value for a classical GSD
"n" no p-value is calculated

The calculated p-values are saved as:
pvalue.r repeated p-value
pvalue.so stage-wise adjusted p-value

etype defines the type of point estimate
"ml" maximum likelihood estimate (ignoring the sequential nature of the design)
"mu" median unbiased estimate (stage-wise lower confidence bound at level 0.5) for a classical GSD
"b" both: maximum likelihood and median unbiased point estimate for a classical GSD
"n" No point estimate is calculated

The calculated point estimates are saved as:
est.ml Maximum likelihood estimate
est.mu Median unbiased estimate

The stage-wise adjusted confidence interval and p-value and the median unbiased point estimate can only be calculated at the stage where the trial stops and is only valid if the stopping rule is met.

The repeated confidence interval and repeated p-value and maximum likelihood estimate can be calculated at every stage of the trial and not just at the stage where the trial stops and is also valid if the stopping rule is not met. For calculating the repeated confidence interval or p-value at any stage of the trial the user has to specify the outcome GSDo in the object GSTobj (see example below).

Value

An object of class GSTobj, is basically a list with the elements

cb.so confidence bound based on the stage-wise ordering
cb.r repeated confidence bound
pvalue.so stage-wise adjusted p-value
pvalue.r repeated p-value
est.ml maximum likelihood estimate
est.mu median unbiased point estimate
GSD
K number of stages
al alpha (type I error rate)
a lower critical bounds of group sequential design (are currently always set to -8)
b upper critical bounds of group sequential design
t vector with cumulative information fraction
SF spending function (for details see below)
phi parameter of spending function when SF=3 or 4 (for details see below)
alab alpha-absorbing parameter values of group sequential design
als alpha-values ''spent'' at each stage of group sequential design
Imax maximum information number
delta effect size used for planning the primary trial
GSDo
T stage where trial stops
z z-statistic at stage where trial stops

Note

SF defines the spending function.

SF = 1 O'Brien and Fleming type spending function of Lan and DeMets (1983)

SF = 2 Pocock type spending function of Lan and DeMets (1983)

SF = 3 Power family (c_α* t^phi). phi must be greater than 0.

SF = 4 Hwang-Shih-DeCani family.(1-e^{-phi t})/(1-e^{-phi}), where phi cannot be 0.

A value of SF=3 corresponds to the power family. Here, the spending function is t^{phi}, where phi must be greater than 0. A value of SF=4 corresponds to the Hwang-Shih-DeCani family, with the spending function (1-e^{-phi t})/(1-e^{-phi}), where phi cannot be 0.

If a path is specified for print.pdf, all must be changed to /. If a filename is specified the ending of the file must be (.pdf).

In the current version a should be set to rep(-8,K)

Author(s)

Niklas Hack niklas.hack@meduniwien.ac.at and Werner Brannath werner.brannath@meduniwien.ac.at

See Also

GSTobj, print.GSTobj, plot.GSTobj, summary.GSTobj

Examples


GSD=plan.GST(K=4,SF=1,phi=0,alpha=0.025,delta=6,pow=0.8,compute.alab=TRUE,compute.als=TRUE)

GST<-as.GST(GSD=GSD,GSDo=list(T=2, z=3.1))

GST
plot(GST)

GST<-summary(GST)
plot(GST)

##The repeated confidence interval, p-value and maximum likelihood estimate 
##at the earlier stage T=1 where the trial stopping rule is not met.

summary(as.GST(GSD,GSDo=list(T=1,z=0.7)),ctype="r",ptype="r",etype="ml")

## Not run: 
##If e.g. the stage-wise adjusted confidence interval is calculated at this stage, 
##the function returns an error message

summary(as.GST(GSD,GSDo=list(T=1,z=0.7)),ctype="so",etype="mu")
## End(Not run)

[Package AGSDest version 1.0 Index]