mosaic (version 0.9-0)

CIsim: Compute confidence intervals from (multiple) simulated data sets

Description

This function automates the calculation of coverage rates for exploring the robustness of confidence interval methods.

Usage

CIsim(n, samples = 100, rdist = rnorm, args = list(), estimand = 0,
  conf.level = 0.95, method = t.test, method.args = list(),
  interval = function(x) {     do.call(method, c(list(x, conf.level =
  conf.level), method.args))$conf.int }, estimate = function(x) {    
  do.call(method, c(list(x, conf.level = conf.level), method.args))$estimate },
  verbose = TRUE)

Arguments

n

size of each sample

samples

number of samples to simulate

rdist

function used to draw random samples

args

arguments required by rdist

estimand

true value of the parameter being estimated

conf.level

confidence level for intervals

method

function used to compute intervals. Standard functions that produce an object of class htest can be used here.

method.args

arguments required by method

interval

a function that computes a confidence interval from data. Function should return a vector of length 2.

estimate

a function that computes an estimate from data

verbose

print summary to screen?

Value

• A data frame with variables lower, upper, estimate, cover ('Yes' or 'No'), and sample is returned invisibly. See the examples for a way to use this to display the intervals graphically.

Examples

CIsim(10,1000)    # 1000 95\% intervals using t.test; population is N(0,1)
CIsim(10,1000, rdist=rexp, estimand=1)    # this time population is Exp(1)
if (require(Hmisc)) {
  xYplot(Cbind(estimate,lower,upper) ~ sample,
    data=CIsim(10,100, rdist=rexp, estimand=1),
    par.settings=col.mosaic(), groups=cover)
  ladd(panel.abline(h=1))
  xYplot( Cbind(estimate,lower,upper) ~ sample,
  data=CIsim(10, 100, rdist=rbinom, args=list(size=1,prob=.5),
       estimand = .5, method = prop.test),
    par.settings=col.mosaic(), groups=cover)
}