AceLavaanGroup: A simple multiple-group ACE model with the 'lavaan' package.
In NlsyLinks: Utilities and kinship information for research with the NLSY

Description Usage Arguments Details Value Note Author(s) References See Also Examples

This function uses the lavaan package to estimate a univariate ACE model, using multiple groups. Each group has a unique value of R (i.e., the Relatedness coefficient).

1	AceLavaanGroup(dsClean, estimateA=TRUE, estimateC=TRUE, printOutput=FALSE)

`dsClean`	The `data.frame` containing complete cases for the `R` groups to be included in the estimation.
`estimateA`	Should the A variance component be estimated? A^2 represents the proportion of variability due to a shared genetic influence.
`estimateC`	Should the C variance component be estimated? C^2 represents the proportion of variability due to a shared environmental influence.
`printOutput`	Indicates if the estimated parameters and fit statistics are printed to the console.

The variance component for E is always estimated, while the A and C estimates can be fixed to zero (when estimateA and/or estimateC are set to FALSE).

An AceEstimate object.

Currently, the variables in dsClean must be named O1, O2 and R; the letter ‘O’ stands for Outcome. This may not be as restrictive as it initially seems, because dsClean is intented to be produced by CleanSemAceDataset. If this is too restrictive for your uses, we'd like to here about it (please email wibeasley at hotmail period com).

Will Beasley

The lavaan package is developed by Yves Rosseel at Ghent University. Three good starting points are the package home page (http://lavaan.ugent.be/), the documentation (http://cran.r-project.org/web/packages/lavaan/) and the JSS paper.

Rosseel, Yves (2012), lavaan: An R Package for Structural Equation Modeling. Journal of Statistical Software, 48, (2), 1-36.

CleanSemAceDataset. Further ACE model details are discussed in our package's vignettes.

library(NlsyLinks) #Load the package into the current R session.
dsLinks <- Links79PairExpanded #Start with the built-in data.frame in NlsyLinks
dsLinks <- dsLinks[dsLinks$RelationshipPath=='Gen2Siblings', ] #Use only Gen2 Siblings (NLSY79-C)

oName_S1 <- "MathStandardized_S1" #Stands for Outcome1
oName_S2 <- "MathStandardized_S2" #Stands for Outcome2

dsGroupSummary <- RGroupSummary(dsLinks, oName_S1, oName_S2)
dsClean <- CleanSemAceDataset(dsDirty=dsLinks, dsGroupSummary, oName_S1, oName_S2)

ace <- AceLavaanGroup(dsClean)
ace

#Should produce:
# [1] "Results of ACE estimation: [show]"
#     ASquared     CSquared     ESquared    CaseCount
#    0.6681874    0.1181227    0.2136900 8390.0000000

library(lavaan) #Load the package to access methods of the lavaan class.
GetDetails(ace)

#Exmaine fit stats like Chi-Squared, RMSEA, CFI, etc.
fitMeasures(GetDetails(ace)) #The function 'fitMeasures' is defined in the lavaan package.

#Examine low-level details like each group's individual parameter estimates and standard errors.
summary(GetDetails(ace))

#Extract low-level details. This may be useful when programming simulations.
inspect(GetDetails(ace), what="converged") #The lavaan package defines 'inspect'.
inspect(GetDetails(ace), what="coef")