bcrm (version 0.4.7)

Posterior: Calculate posterior distribution of CRM model parameter(s)

Description

Given the prior, functional form of the dose-toxicity model, and data, this function returns the posterior distribution (via either MCMC samples, or posterior summaries) of the CRM model parameter(s)

Usage

Posterior.exact(tox,notox,sdose,ff,prior.alpha)
Posterior.rjags(tox,notox,sdose,ff,prior.alpha,burnin.itr,production.itr)
Posterior.BRugs(tox,notox,sdose,ff,prior.alpha,burnin.itr,production.itr)
Posterior.R2WinBUGS(tox,notox,sdose,ff,prior.alpha,burnin.itr,production.itr
  ,bugs.directory)

Arguments

tox

A vector of length k listing the number of patients who have experienced the outcome (toxicity) at each dose level 1,...,k. If missing, then it is assumed that no data have thus far been collected.

notox

A vector of length k listing the number of patients who have not experienced the outcome (toxicity) at each dose level 1,...,k. If missing, then it is assumed that no data have thus far been collected.

sdose

A vector of length k listing the standardised doses used in the CRM model.

ff

A string indicating the functional form of the dose-response curve. Options are

ht

1-parameter hyperbolic tangent

logit1

1-parameter logistic

power

1-parameter power

logit2

2-parameter logistic

prior.alpha

A list of length 3 containing the distributional information for the prior. The first element is a number from 1-4 specifying the type of distribution. Options are

1. Gamma(a,b), where a=shape, b=scale: mean=a*b, variance=a*b*b

2. Uniform(a,b), where a=min, b=max

3. Lognormal(a,b), where a=mean on the log scale, b=standard deviation on the log scale

4. Bivariate Lognormal(a,b), where a=mean vector on the log scale, b=Variance-covariance matrix on the log scale. This prior should be used only in conjunction with a two-parameter logistic model.

The second and third elements of the list are the parameters a and b, respectively.

burnin.itr

Number of burn-in iterations (default 2000).

production.itr

Number of production iterations (default 2000).

bugs.directory

Directory that contains the WinBUGS executable if method="R2WinBUGS". Defaults to "C:/Program Files/WinBUGS14/".

Details

Posterior.exact produces posterior summary statistics of the CRM model parameter(s), and probabilities of toxicity at the dose levels using exact Bayesian computation (integration) of the posterior distribution. If Posterior.BRugs or Posterior.R2WinBUGS is specified, then posterior samples of the CRM model parameter(s) is returned by the function.

References

Sweeting M., Mander A., Sabin T. bcrm: Bayesian Continual Reassessment Method Designs for Phase I Dose-Finding Trials. Journal of Statistical Software (2013) 54: 1--26. http://www.jstatsoft.org/article/view/v054i13

See Also

bcrm

Examples

# NOT RUN {
## Dose-escalation cancer trial example as described in Neuenschwander et al 2008.
## Pre-defined doses
dose<-c(1,2.5,5,10,15,20,25,30,40,50,75,100,150,200,250)
## Pre-specified probabilities of toxicity
## [dose levels 11-15 not specified in the paper, and are for illustration only]
p.tox0<-c(0.010,0.015,0.020,0.025,0.030,0.040,0.050,0.100,0.170,0.300,0.400,0.500,0.650
  ,0.800,0.900)
## Data from the first 5 cohorts of 18 patients
tox<-c(0,0,0,0,0,0,2,0,0,0,0,0,0,0,0)
notox<-c(3,4,5,4,0,0,0,0,0,0,0,0,0,0,0)
## Target toxicity level
target.tox<-0.30
## Lognormal prior
prior.alpha<-list(3,0,1.34^2)
## Power functional form
ff<-"power"
## Standardised doses
sdose<-find.x(ff,p.tox0,alpha=1)

## Posterior distribution of the model parameter using exact computation
post.exact<-Posterior.exact(tox,notox,sdose,ff,prior.alpha)
print(post.exact)

## Posterior distribution of the model parameter using rjags
post.rjags<-Posterior.rjags(tox,notox,sdose,ff,prior.alpha
  ,burnin.itr=2000,production.itr=2000)
print(mean(post.rjags))
hist(post.rjags)

## Posterior distribution of the model parameter using BRugs (Windows and i386 Linux only)
if(Sys.info()["sysname"] %in% c("Windows","Linux")){
	post.BRugs<-Posterior.BRugs(tox,notox,sdose,ff,prior.alpha
	  ,burnin.itr=2000,production.itr=2000)
	print(mean(post.BRugs))
	hist(post.BRugs)
	}
# }