getPars: Calculate location and scale parameters

Description

Given the baseline, the maximum effect and the standardized model parameters this function calculates the location and scale parameters in the model function using the maximum approach, see Pinheiro et al. (2006) for the basic idea.

Usage

getPars(model, doses, initEstim, base, maxEff,
        off = 0.1 * max(doses), scal = 1.2 * max(doses))

Arguments

model

A character string with the model name. Built-in models have their full parameterization derived internally. For user-defined models, it is assumed that a function named as "Par" appended to end of model name exists (e.g., for model = "cubic", it is assumed that there is function "cubicPar" that calculates the necessary parameters; this function is assumed to have arguments "doses", "initEstim", "base", and "maxEff", in that order (see below for an example).

doses

Doses to be used in design

initEstim

Vector of guesstimates

base

Expected baseline effect

maxEff

Expected maximum change from baseline

off

Offset parameter for the linear in log model (default 1).

scal

Scale parameter for the beta model (default: 20 perc. larger than maximum dose).

Value

Vector containing all model parameters.

References

Pinheiro, J. C., Bornkamp, B. and Bretz, F. (2006). Design and analysis of dose finding studies combining multiple comparisons and modeling procedures, Journal of Biopharmaceutical Statistics, 16, 639--656

Examples

Run this code

# NOT RUN {
doses <- c(0, 10, 25, 50, 100, 150)                                        
getPars("emax", doses, 25, 0, 0.4)                                         
getPars("logistic", doses, c(50, 10.88111), 0, 0.4) # compare JBS 16, p.650
getPars("betaMod", doses, initEstim = c(0.33, 2.31), base = 0,             
        maxEff = 0.4)                                                      
#example for user model                                                    
userMod <- function(dose, e0, eMax, ed50, h){                             
  e0 + eMax * ( dose^h / (ed50^h + dose^h) )                               
}                                                                          
# function to return location and scale parameters                         
userModPar <- function(dose, initEstim, base, maxEff){                    
  # function to get linear parameters                                      
  # ed50 parameter assumed to be first in initEstim                        
  ed50 <- initEstim[1]                                                     
  h <- initEstim[2]                                                        
  dmax <- max(dose)                                                        
  emax <- maxEff*(ed50^h+dmax^h)/dmax^h                                    
  c(base, emax, initEstim)                                                 
}                                                                          
getPars("userMod", doses, initEstim = c(50,2), base = 0, maxEff = 1) 
# }