HWPower {HardyWeinberg}

Compute the power of a test for Hardy-Weinberg equilibrium.

Description

HWPower is a function that computes the power of a test for Hardy-Weinberg equilibrium.

Usage

HWPower(n = 100, nA = 100, pA = 0.5, y = c(AA=25,AB=50,BB=25),
alpha = 0.05, theta = 4, f = NULL, test = "exact",
alternative = "two.sided", pvaluetype = "selome", cc = 0.5)

Arguments

n

The sample size

nA

The minor allele count

pA

The minor allele frequency

y

A sample of genotype counts (AA,AB,BB)

alpha

The significance level (0.05 by default)

theta

The degree of disequilibrium (theta = 4 is equilibrium, theta > 4 is heterozygote excess, theta < 4 is heterozygote dearth)

f

The inbreeding coefficient. Overrules theta if specified.

test

The type of test for which power is to be computed. Can be "exact" (default) or "chisq" (chi-square)

alternative

The nature of the alternative hypothesis ("two.sided" (default), "greater" or "less")

pvaluetype

The type of p-value used in an exact test ("selome", "dost" or "midp")

cc

Continuity correction parameter for the chi-square test (0.5 by default)

Details

HWPower uses the Levene-Haldane distribution (distribution of the number of heterzygotes given the minor allele count) for computing power.

HWPower can be used in three different way. In principle, the power is calcuted on the basis of the sample size (n) and the minor allele count (nA). Alternatively, the user may specifiy sample size (n) and minor allele frequency (pA). Finally, power can also be calculated directly from a sample of genotype counts. In that case the calculated power is the power for a sample of the given sample size and minor allele count. The three ways to use HWPower are illustrated in the example section.

Value

if test = "exact" the power of the exact test is computed for the given significance level and minor allele count.

if test = "chisq" the power of the chi-square test is computed for the given significance level and minor allele count.

Author(s)

Jan Graffelman ([e-mail])

References

Graffelman, J. and Moreno, V. (2013) The Mid p-value in exact tests for Hardy-Weinberg proportions. Statistical Applications in Genetics and Molecular Biology 12(4): 433-448.

Graffelman, J. (2015) Exploring Diallelic Genetic Markers: The HardyWeinberg Package. Journal of Statistical Software 64(3): 1-23. doi:10.18637/jss.v064.i03.

See Also

HWExact

Examples

pw.chisq <- HWPower(n=100,nA=100,alpha=0.05,test="chisq",theta=16)
print(pw.chisq)
pw.exact <- HWPower(n=100,nA=100,alpha=0.05,test="exact",theta=16,pvaluetype="selome")
print(pw.exact)
pw.exact <- HWPower(n=100,nA=100)
print(pw.exact)
pw.exact <- HWPower(n=100,pA=0.5)
print(pw.exact)
pw.exact <- HWPower(y=c(AA=25,AB=50,BB=25))
print(pw.exact)