close_to_normal: This is a CLT-linked run-time control.

In gwrpvr: Genome-Wide Regression P-Value (Gwrpv)

Description

If the number of observations is large enough that a normality approximation holds for the y average across the major homozygote subsample, then the code skips the time-consuming loop over n0, n1 and n2 and and uses the normal approximation for the average y for the major homozygote subsample. The remaining loop is only over n1 and n2. The only new input/output variables are input lognearnorm (the magnitude of maximum allowed tolerance (in log 10 format) for the sum of squared deviation of skewness and kurtosis from their normal values and output stopiter (a zero if the code does not mandate a stop to the iterative estimation and a one if it does). The input variable lognearnorm has a default value set so that users only have to enter it if they want to over-ride the default value.

Usage

close_to_normal(totnobs, n0, n1, n2, pa, pb, mua, mub, siga, sigb, beta,
  nearnorm)

Arguments

totnobs	the sum of n0, n1, n2
n0	the major allele homozygotes
n1	the major allele heterozygotes
n2	the minor allele zygotes
pa	parameter of the mixture distribution, a real number between zero and one with pa+pb=1
pb	parameter of the mixture distribution, a real number between zero and one with pa+pb=1
mua	parameter of the mixture distribution, can be any real number
mub	parameter of the mixture distribution, can be any real number
siga	parameter of the mixture distribution, can be any real number
sigb	parameter of the mixture distribution, can be any real number
beta	the beta from the regression being tested
nearnorm	must be in log base 10 format, with default value set to -5

Value

list(skewbeta = skewbeta, kurtbeta = kurtbeta, sigbeta = sigbeta, skipiter = skipiter)

gwrpvr documentation built on May 2, 2019, 7:59 a.m.