AtomicSim.US: Atomic simulation of AUC coverage experiment with...

Description Usage Arguments Details Value Examples

View source: R/AtomicSim.US.R

Description

Atomic simulation of AUC coverage experiment with undersmoothing

Usage

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
AtomicSim.US(
  distribution,
  n,
  AUC,
  theta_squared,
  var_eq,
  ME_var_ratio,
  R = 1000,
  US_ratio = c(0.5, 0.25)
)

Arguments

distribution

one of the following strings : "normal", "gamma.skewed", "gamma.symmetrical" , "log.normal", "bimodal1", "bimodal2", "t3", "laplace", "cauchy"

n

number of observations in each group

AUC

the probability P(X>Y) of correctly ranking a pair of random observations (diseased,healthy).

theta_squared

contamination parameter (noise/signal)

var_eq

logical - if TRUE VX=VY=1, if FALSE VX=2VY=2 ratio of the biomarker variances

ME_var_ratio

(defaults to 1) the ration between ME variances, V

R

bootstrap replications (defaults to 1000)

US_ratio

vector of multipliers for the undersmoothing of bootstrap samples (defaults to 0.5,0.25)

Details

This is the doOne function required by simsalapar for parallel simulations.

Value

a vector of the point estimate and percentile CI and also the original observations,all concatenated in a named vector.

Examples

1
AtomicSim("gamma.skewed",250,0.8,0.5,TRUE,1,R=1000)

blebedenko/thescript2 documentation built on Dec. 19, 2021, 9:53 a.m.