poolbinom.lrt: Calculate LRT based confidence interval for binomial...

In hoehleatsu/binomSamSize: Confidence Intervals and Sample Size Determination for a Binomial Proportion under Simple Random Sampling and Pooled Sampling

Calculate LRT based confidence interval for binomial proportion for pooled samples

Description

Calculate LRT based confidence interval for the Bernoulli proportion of k\cdot n individuals, which are pooled into n pools each of size k. Observed is the number of positive pools x.

Usage

poolbinom.lrt(x, k, n, conf.level=0.95, bayes=FALSE, conf.adj=FALSE)

Arguments

x	Number of positive pools (can be a vector).
k	Pool size (can be a vector).
n	Number of pools (can be a vector).
conf.level	The level of confidence to be used in the confidence interval
bayes	See binom.bayes
conf.adj	See binom.bayes

Details

Compute LRT based intervals for the binomial response X \sim Bin(n, \theta), where \theta = 1 - (1-\pi)^k. As a consequence,

\pi = g(\theta) = 1 - (1-\pi)^{1/k}

.

One then knows that the borders for \pi are just transformations of the borders of theta using the above g(\theta) function.

For further details about the pooling setup see poolbinom.logit.

Value

A data.frame containing the observed proportions and the lower and upper bounds of the confidence interval. The output is similar to the binom.confint function of the binom package

Author(s)

M. Höhle

Examples

binom.lrt(x=0:34,n=34)
poolbinom.lrt(x=0:34,k=1,n=34)
poolbinom.lrt(x=0:34,k=10,n=34)

hoehleatsu/binomSamSize documentation built on March 25, 2024, 10:07 p.m.