gap (version 1.0-8)

twinan90: Classic twin models

Description

Classic twin models

The function also allows for test for normality and Box-Cox transformation. It further allows for output data in FISHER format.

Usage

twinan90(mzdat,dzdat,vname='mzdz',xlamb=1,const=0,vmiss=-9,path=1,
           ped=0,nvar=1,form='((1x,a1,5x,F6.2))')

Arguments

mzdat

two columns of data for MZ twins

dzdat

two columns of data for DZ twins

vname

variable name

xlamb

the Box-Cox transformation parameter

const

constant to be added to each data value

vmiss

missing value indicator

path

path analysis to be conducted

ped

if 1, to output data in FISHER format

nvar

number of variables in the data file

form

the Fortran format for the data file

Value

• The returned values is in a list with two components, while the MLEs are in two system files:
• h2
  The heritability estimate based on 2(rMZ-rDZ), where rMZ and rDZ are the intraclass correlation coefficients for MZ and DZ twin pairs
• seh2
  The standard error for the above statistic
• vname.log
  the diagnostic file
• vname.out
  the output file

References

Williams CJ, Christian JC, Norton JA Jr. (1992) TWINAN90: A FORTRAN program for conducting ANOVA-based and likelihood-based analyses of twin data. Computer Methods and Programs in Biomedicine 38:(2-3):167-176

Williams CJ (1993). On the covariance between parameter estimates in models of twin data. Biometrics. 49(2):557-68

Examples

mzdz <- matrix(scan("mzdz.dat",skip=1),ncol=2,byrow=T)
mzdat <- mzdz[1:131,]
dzdat <- mzdz[132:206,]
twinan90(mzdat,dzdat,xlamb=2)