devel/Dual endpoint separate model 27012015.R

In 0liver0815/onc-crmpack-test: Object-Oriented Implementation of CRM Designs

library(crmPack)
mySize<-CohortSizeConst(size=3)
myIncrements1 <-IncrementsRelative(intervals=c(2),increments=c(2))

data2<-DataDual(doseGrid=c(seq(2,10, by=1),12,15,20,24,30,40,50,60,70,80,90,100,110,120))

data4<-DataDual(x=c(9,9,9,9,9),
                y=c(0,0,0,0,0),
                w=c(0.404,1.232,0.328,-1.040,-1.588),
                doseGrid=c(seq(2,10, by=1),12,15,20,24,30,40,50,60,70,80,90,100,110,120))

#efficacy model

Effmodel<-Effloglog(Eff=c(0.05,1.0),Effdose=c(2,120),nu=c(a=0.001,b=0.001),data=data2)


newEffmodel <- update(object=Effmodel,data=data4)



#dlt model
DLTmodel<-LogisticIndepBeta(binDLE=c(0.0713, 0.9507), DLEweights=c(3,3),DLEdose=c(2,120),data=data2)


data3<-Data(
x=c(9,9,9,9,9),
y=c(0,0,0,0,0),
doseGrid=c(seq(2,10, by=1),12,15,20,24,30,40,50,60,70,80,90,100,110,120))

#efficacy model
#Modal estimates
newDLTmodel <-update(object=DLTmodel, data=data3)


#*****
GainNextBest<-NextBestMaxGain(DLEDuringTrialtarget=0.35, DLEEndOfTrialtarget=0.3)
doseRecGain <- nextBest(GainNextBest,
                    doselimit=max(data4@doseGrid),
                        model=newDLTmodel,
                        Effmodel=newEffmodel,
                        data=data4)
doseRecGain$plot
doseRecGain


#stopping
myStopping7 <- StoppingGstarCIRatio(targetRatio=5, 
                                    targetEndOfTrial=0.3)

myStopping8 <- myStopping7 | StoppingMinPatients(72)
stopTrial(stopping=myStopping7,dose=doseRecGain$nextdose,model=newDLTmodel,
          data=data4, Effmodel=newEffmodel)


stopTrial(stopping=myStopping8,dose=doseRecGain$nextdose,model=newDLTmodel,
          data=data4, Effmodel=newEffmodel)


## check how lapply works with missing arguments
myfun <- function(x, a, b)
{
  if(missing(b))
  {
    x + a
  } else {
    x + a + b
  }
}
test <- function(x, a, b)
{
lapply(x, 
       myfun,
       a,
       b)
}
test(1:5, a=3, b=2)