inst/models/passing/DefinitionMeans.R

In OpenMx/OpenMx: Extended Structural Equation Modelling

#
#   Copyright 2007-2018 by the individuals mentioned in the source code history
#
#   Licensed under the Apache License, Version 2.0 (the "License");
#   you may not use this file except in compliance with the License.
#   You may obtain a copy of the License at
# 
#        http://www.apache.org/licenses/LICENSE-2.0
# 
#   Unless required by applicable law or agreed to in writing, software
#   distributed under the License is distributed on an "AS IS" BASIS,
#   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
#   See the License for the specific language governing permissions and
#   limitations under the License.

require(OpenMx)

#Definition Variable Test 3
#Author: Mike Neale
#Date: July 29 2009

#This script is used to test the definition variable functionality in OpenMx
#The definition variable in this example is dichotomous, and describes two different groups
#These two groups are measured on two variables, x and y
#The group with a definition value of 1 has means of 1 and 2 for x and y
#The group with a definition value of 0 has means af zero for x and y 
#The definition variable is then used to define a mean deviation of the group with definition value 1

#make some data!
set.seed(200)
n = 500

Sigma <- matrix(c(1,.5,.5,1),2,2)
group1<-mvtnorm::rmvnorm(n=n, c(1,2), Sigma)
group2<-mvtnorm::rmvnorm(n=n, c(0,0), Sigma)

#put them both together, add a definition variable, and make an selection variables object
y<-rbind(group1,group2)
dimnames(y)[2]<-list(c("x","y"))
def<-rep(c(1,0),each=n)
selvars<-c("x","y")

if (0) {
  #write data to a file for the mx script to read (not necessary for running in R)
  write.table(cbind(y,def),file="temp-files/xydefmeans.rec",col.names=F,row.names=F)
}

# Three covariance model matrices: 
#  "cov" for the zero relationship group
#  "def" for the definition variable, 
#   and "beta" for estimating difference between groups' covariances
# One common mean vector, "M"

#define the model, including a FIML objective function, which will optimize the matrix S
model<-mxModel("model", mxFitFunctionML(),mxExpectationNormal("Sigma", "Mu", selvars), 
				mxData((data.frame(y,def)), type="raw"),
				
				mxMatrix("Symm", nrow=2, ncol=2, free=TRUE, values=c(1, 0, 1), name="Sigma"),
					
				mxMatrix("Full", nrow=1, ncol=2, free=TRUE, values=c(0, 0),
					dimnames=list(NULL, selvars), name="beta"),
				mxMatrix("Full", nrow=1, ncol=2, free=FALSE, labels=c("data.def"),
					dimnames=list(NULL, selvars), name="def"),
				mxMatrix("Full", nrow=1, ncol=2, free=TRUE, 
					dimnames=list(NULL, selvars), name = "M"),
				
				mxAlgebra(M+beta*def, name="Mu")
			)

#run the model
run<-mxRun(model)
run$matrices
run$algebras


#Compare OpenMx estimates to summary statistics from raw data, remembering to knock off 1 and 2 from group 1's
# data, so as to estimate variance of combined sample without the mean correction.

# First we compute some summary statistics from the data
ObsCovs <- cov(rbind(group1 - rep(c(1,2), each=n), group2))
ObsMeansGroup1 <- c(mean(group1[,1]), mean(group1[,2]))
ObsMeansGroup2 <- c(mean(group2[,1]), mean(group2[,2]))

# Second we extract the parameter estimates and matrix algebra results from the model
Sigma<-run$matrices$Sigma$values
M<-run$matrices$M$values
beta<-run$matrices$beta$values

# Third, we check to see if things are more or less equal
omxCheckCloseEnough(ObsCovs,Sigma,.01)
omxCheckCloseEnough(ObsMeansGroup1,as.vector(M+beta),.001)
omxCheckCloseEnough(ObsMeansGroup2,as.vector(M),.001)