> rm(list=ls())
> 
> #load packages
> library(readr)
> library(plyr)
> library(psych)
> library("OpenMx")
> ### Specify optimizer
> mxOption(NULL, "Default optimizer", "SLSQP")
> 
> #load data
> mtfs <- read.csv("/Users/zelle063/Desktop/Vrieze Lab/GEDI_Cleaning/CleanData/SUDEV.csv")
> 
> #create cohort variables
> mtfs$OC[mtfs$IDAC==17] <- 1
> mtfs$OC[mtfs$IDAC==11] <- 0
> mtfs$YC[mtfs$IDAC==17] <- 0
> mtfs$YC[mtfs$IDAC==11 & mtfs$IDES == 9] <- 1
> mtfs$YC[mtfs$IDAC==11 & mtfs$IDES != 9] <- 0
> 
> mtfs$ES[mtfs$IDES == 9] <- 0
> mtfs$ES[mtfs$IDES != 9] <- 1
> 
> 
> 
> 
> ##reshape to wide format
> mtfs <- subset(mtfs, select=c("IDYRFAM", "IDSC", "IDSEX", "ZYGOSITY",
+                               "AlcFreq14",     "AlcFreq17",     "AlcFreq20",     "AlcFreq24",     "AlcFreq29",     "AlcFreq35",    "AlcFreq45",
+                               "AlcQuant14",    "AlcQuant17",    "AlcQuant20",    "AlcQuant24",    "AlcQuant29",    "AlcQuant35",   "AlcQuant45",
+                               "CPD_TOB_age14", "CPD_TOB_age17", "CPD_TOB_age20", "CPD_TOB_age24", "CPD_TOB_age29", "CPD_TOB_age35","CPD_TOB_age45",
+                               "MJ_FREQ_14",    "MJ_FREQ_17",    "MJ_FREQ_20",    "MJ_FREQ_24",    "MJ_FREQ_29",    "MJ_FREQ_35", "MJ_FREQ_45", "OC", "YC"))
> 
> mtfs.w <- reshape(mtfs, v.names=c("AlcFreq14",     "AlcFreq17",     "AlcFreq20",     "AlcFreq24",     "AlcFreq29",     "AlcFreq35",    "AlcFreq45",
+                                   "AlcQuant14",    "AlcQuant17",    "AlcQuant20",    "AlcQuant24",    "AlcQuant29",    "AlcQuant35",   "AlcQuant45",
+                                   "CPD_TOB_age14", "CPD_TOB_age17", "CPD_TOB_age20", "CPD_TOB_age24", "CPD_TOB_age29", "CPD_TOB_age35","CPD_TOB_age45",
+                                   "MJ_FREQ_14",    "MJ_FREQ_17",    "MJ_FREQ_20",    "MJ_FREQ_24",    "MJ_FREQ_29",    "MJ_FREQ_35", "MJ_FREQ_45"), timevar="IDSC", idvar=c("IDYRFAM", "IDSEX", "OC", "YC"), direction="wide", sep="")
> 
> 
> 
> ########################
> ### Biometric Models ###
> ########################
> ###
> ### Select variables
> Vars <- c("AlcFreq24",  "AlcQuant24",  "CPD_TOB_age24", "MJ_FREQ_24")
> 
> nv <- length(Vars)
> ntv <- nv*2
> selVars <- paste(Vars, c(rep(0,nv), rep(1,nv)), sep="")
> 
> DefVars <- c("OC", "YC")
> 
> ### Select Data for Analysis
> mtfsMzDatam    <- subset(mtfs.w, ZYGOSITY==1 & IDSEX == 1, select=c(selVars, DefVars))
> mtfsDzDatam    <- subset(mtfs.w, ZYGOSITY==2 & IDSEX == 1, select=c(selVars, DefVars))
> 
> mtfsMzDataTm <- mtfsMzDatam[rowSums(is.na(mtfsMzDatam)) != ncol(mtfsMzDatam),]
> mtfsDzDataTm <- mtfsDzDatam[rowSums(is.na(mtfsDzDatam)) != ncol(mtfsDzDatam),]
> 
> mtfsDataMZm    <- mxData( observed=mtfsMzDataTm, type="raw" )
> mtfsDataDZm    <- mxData( observed=mtfsDzDataTm, type="raw" )
> 
> mtfsMzDataf    <- subset(mtfs.w, ZYGOSITY==1 & IDSEX == 2, select=c(selVars, DefVars))
> mtfsDzDataf    <- subset(mtfs.w, ZYGOSITY==2 & IDSEX == 2, select=c(selVars, DefVars))
> 
> mtfsMzDataTf <- mtfsMzDataf[rowSums(is.na(mtfsMzDataf)) != ncol(mtfsMzDataf),]
> mtfsDzDataTf <- mtfsDzDataf[rowSums(is.na(mtfsDzDataf)) != ncol(mtfsDzDataf),]
> 
> mtfsDataMZf   <- mxData( observed=mtfsMzDataTf, type="raw" )
> mtfsDataDZf   <- mxData( observed=mtfsDzDataTf, type="raw" )
> 
> 
> ### ------------------------------------------------------------------------------
> ### Common Pathway ACE model
> ### ------------------------------------------------------------------------------
> nl <- 1 # number of latent factors
> # Matrices ac, cc, and ec to store a, c, and e path coefficients for latent phenotype(s)
> Xm <- mxMatrix(type="Lower", nrow=nl, ncol=nl, free=TRUE, 
+                values=c(.6), 
+                labels=c("x11m"),
+                name="Xm", byrow = T)
> 
> Ym <- mxMatrix(type="Lower", nrow=nl, ncol=nl, free=TRUE, 
+                values=c(.6), 
+                labels=c("y11m"),
+                name="Ym", byrow = T)
> 
> Zm <- mxMatrix(type="Lower", nrow=nl, ncol=nl, free=TRUE, 
+                values=c(.6), 
+                labels=c("z11m"),
+                name="Zm", byrow = T )
> 
> 
> Xf <- mxMatrix(type="Lower", nrow=nl, ncol=nl, free=TRUE, 
+                values=c(.6), 
+                labels=c("x11f"),
+                name="Xf", byrow = T)
> 
> Yf <- mxMatrix(type="Lower", nrow=nl, ncol=nl, free=TRUE, 
+                values=c(.6), 
+                labels=c("y11f"),
+                name="Yf", byrow = T)
> 
> Zf <- mxMatrix(type="Lower", nrow=nl, ncol=nl, free=TRUE, 
+                values=c(.6), 
+                labels=c("z11f"),
+                name="Zf", byrow = T )
> 
> 
> Alm <- mxAlgebra(Xm %*% t(Xm), name="Alm")
> Clm <- mxAlgebra(Ym %*% t(Ym), name="Clm")
> Elm <- mxAlgebra(Zm %*% t(Zm), name="Elm")
> 
> Alf <- mxAlgebra(Xf %*% t(Xf), name="Alf")
> Clf <- mxAlgebra(Yf %*% t(Yf), name="Clf")
> Elf <- mxAlgebra(Zf %*% t(Zf), name="Elf")
> 
> A <- c(TRUE, TRUE, F,    F,
+        TRUE, TRUE, F,    F,
+        F,    F,    TRUE, F,
+        F,    F,    F,    TRUE)
> 
> SpecificFree <- matrix(A, nrow=nv, ncol=nv, byrow=T)
> 
> SpecificVals <- SpecificFree
> SpecificVals[which(SpecificVals==TRUE)] <- .4
> SpecificVals[which(SpecificVals==FALSE)] <- 0
> diag(SpecificVals) <- .5
> 
> labelsData <- c("AFasm", "ALCasm", NA, NA,
+                 "ALCasm", "AQasm", NA, NA,
+                 NA, NA, "CPDasm", NA,
+                 NA, NA, NA, "MJasm")
> labelsAsm <- matrix(labelsData, nrow=nv, ncol=nv, byrow=T)
> 
> labelsCsm <- gsub("asm", "csm", labelsAsm) 
> labelsEsm <- gsub("asm", "esm", labelsAsm)
> labelsAsf <- gsub("asm", "asf", labelsAsm)
> labelsCsf <- gsub("asm", "csf", labelsAsm)
> labelsEsf <- gsub("asm", "esf", labelsAsm)
> 
> # Matrices as, cs, and es to store a, c, and e path coefficients for specific factors
> pathAsm <- mxMatrix(type="Symm", nrow=nv, ncol=nv, free=SpecificFree, values=SpecificVals, 
+                     labels=labelsAsm, name="pathAsm", byrow = T )
> 
> pathCsm <- mxMatrix(type="Symm", nrow=nv, ncol=nv, free=SpecificFree, values=SpecificVals, 
+                     labels=labelsCsm, name="pathCsm", byrow = T )
> 
> pathEsm <- mxMatrix(type="Symm", nrow=nv, ncol=nv, free=SpecificFree, values=SpecificVals, 
+                     labels=labelsEsm, name="pathEsm", byrow = T )
> 
> pathAsf <- mxMatrix(type="Symm", nrow=nv, ncol=nv, free=SpecificFree, values=SpecificVals, 
+                     labels=labelsAsf, name="pathAsf", byrow = T )
> 
> pathCsf <- mxMatrix(type="Symm", nrow=nv, ncol=nv, free=SpecificFree, values=SpecificVals, 
+                     labels=labelsCsf, name="pathCsf", byrow = T )
> 
> pathEsf <- mxMatrix(type="Symm", nrow=nv, ncol=nv, free=SpecificFree, values=SpecificVals, 
+                     labels=labelsEsf, name="pathEsf", byrow = T )
> 
> # Matrix and Algebra for constraint on variance of latent phenotype
> CovarLPm <- mxAlgebra( expression= Alm + Clm + Elm, name="CovarLPm" )
> VarLPm <- mxAlgebra( expression= diag2vec(CovarLPm), name="VarLPm" )
> Unitm <- mxMatrix( type="Unit", nrow=nl, ncol=1, name="Unitm")
> varLP1m <- mxConstraint( expression=VarLPm == Unitm, name="varLP1m")
> 
> CovarLPf <- mxAlgebra( expression= Alf + Clf + Elf, name="CovarLPf" )
> VarLPf <- mxAlgebra( expression= diag2vec(CovarLPf), name="VarLPf" )
> Unitf <- mxMatrix( type="Unit", nrow=nl, ncol=1, name="Unitf")
> varLP1f <- mxConstraint( expression=VarLPf == Unitf, name="varLP1f")
> 
> 
> ### Matrix f for factor loadings on latent phenotype
> pathFlm <- mxMatrix(type="Full", nrow=nv, ncol=nl, free=T, values=c(.6, .6, .6, .6), labels=c("flAFm", "flAQm", "flCPDm", "flMJm"), name="pathFlm", byrow = T)
> pathFlf <- mxMatrix(type="Full", nrow=nv, ncol=nl, free=T, values=c(.6, .6, .6, .6), labels=c("flAFf", "flAQf", "flCPDf", "flMJf"), name="pathFlf", byrow = T)
> 
> 
> # Matrices A, C, and E compute variance components
> covAm <- mxAlgebra( expression=pathFlm %*% Alm %*% t(pathFlm) + pathAsm %*% t(pathAsm), name="Am" )
> covCm <- mxAlgebra( expression=pathFlm %*% Clm %*% t(pathFlm) + pathCsm %*% t(pathCsm), name="Cm" )
> covEm <- mxAlgebra( expression=pathFlm %*% Elm %*% t(pathFlm) + pathEsm %*% t(pathEsm), name="Em" )
> 
> covAf <- mxAlgebra( expression=pathFlf %*% Alf %*% t(pathFlf) + pathAsf %*% t(pathAsf), name="Af" )
> covCf <- mxAlgebra( expression=pathFlf %*% Clf %*% t(pathFlf) + pathCsf %*% t(pathCsf), name="Cf" )
> covEf <- mxAlgebra( expression=pathFlf %*% Elf %*% t(pathFlf) + pathEsf %*% t(pathEsf), name="Ef" )
> 
> covPm <- mxAlgebra( expression=Am+Cm+Em, name="Vm" )
> matIm <- mxMatrix( type="Iden", nrow=nv, ncol=nv, name="matIm")
> invSDm <- mxAlgebra( expression=solve(sqrt(matIm*Vm)), name="iSDm")
> Sflm <- mxAlgebra(expression = iSDm %*% pathFlm, name = "Sflm")
> 
> covPf <- mxAlgebra( expression=Af+Cf+Ef, name="Vf" )
> matIf <- mxMatrix( type="Iden", nrow=nv, ncol=nv, name="matIf")
> invSDf <- mxAlgebra( expression=solve(sqrt(matIf*Vf)), name="iSDf")
> Sflf <- mxAlgebra(expression = iSDf %*% pathFlf, name = "Sflf")
> 
> averageFLm <- mxAlgebra( expression = ((MZm.Sflm[1,1] + MZm.Sflm[2,1] + MZm.Sflm[3,1] + MZm.Sflm[4,1])/4), name="averageFLm")
> averageFLf <- mxAlgebra( expression = ((MZf.Sflf[1,1] + MZf.Sflf[2,1] + MZf.Sflf[3,1] + MZf.Sflf[4,1])/4), name="averageFLf")
> 
> 
> #regress cohort out of factors
> FacMeansm <- mxMatrix("Full",nrow=nl,ncol=1, free=F, values=0, labels=c("mean1m"), name="FacMeansm")
> 
> regCoefYCm <- mxMatrix(type="Full", nrow=nl, ncol=1, free=TRUE, values=.2, labels=c("beta1YCm"), name="regCoefYCm")
> YCCovm <- mxMatrix(type="Full", nrow=1, ncol=1, free=FALSE, labels="data.YC", name="YCm")
> 
> regCoefOCm <- mxMatrix(type="Full", nrow=nl, ncol=1, free=TRUE, values=.2, labels=c("beta1OCm"), name="regCoefOCm")
> OCCovm <- mxMatrix(type="Full", nrow=1, ncol=1, free=FALSE, labels="data.OC", name="OCm")
> 
> FacMeans2m <- mxAlgebra(expression= (FacMeansm+(regCoefYCm%*%YCm)+(regCoefOCm%*%OCm)), name="FacMeans2m")
> meansm    <- mxAlgebra(t(pathFlm %*% FacMeans2m), name="meansm")
> expMeanm <- mxAlgebra(cbind(meansm, meansm), name="expMeanm")
> 
> 
> FacMeansf <- mxMatrix("Full",nrow=nl,ncol=1, free=F, values=0, labels=c("mean1f"), name="FacMeansf")
> 
> regCoefYCf <- mxMatrix(type="Full", nrow=nl, ncol=1, free=TRUE, values=0, labels=c("beta1YCf"), name="regCoefYCf")
> YCCovf <- mxMatrix(type="Full", nrow=1, ncol=1, free=FALSE, labels="data.YC", name="YCf")
> 
> regCoefOCf <- mxMatrix(type="Full", nrow=nl, ncol=1, free=TRUE, values=0, labels=c("beta1OCf"), name="regCoefOCf")
> OCCovf <- mxMatrix(type="Full", nrow=1, ncol=1, free=FALSE, labels="data.OC", name="OCf")
> 
> FacMeans2f <- mxAlgebra(expression= (FacMeansf+(regCoefYCf%*%YCf)+(regCoefOCf%*%OCf)), name="FacMeans2f")
> meansf    <- mxAlgebra(t(pathFlf %*% FacMeans2f), name="meansf")
> expMeanf <- mxAlgebra(cbind(meansf, meansf), name="expMeanf")
> 
> 
> covMZm <- mxAlgebra( expression= rbind( cbind(Am+Cm+Em, Am+Cm),
+                                         cbind(Am+Cm,  Am+Cm+Em)), name="expCovMZm" )
> covDZm <- mxAlgebra( expression= rbind( cbind(Am+Cm+Em, 0.5%x%Am+Cm),
+                                         cbind(0.5%x%Am+Cm, Am+Cm+Em)), name="expCovDZm" )
> 
> covMZf <- mxAlgebra( expression= rbind( cbind(Af+Cf+Ef, Af+Cf),
+                                         cbind(Af+Cf,  Af+Cf+Ef)), name="expCovMZf" )
> covDZf <- mxAlgebra( expression= rbind( cbind(Af+Cf+Ef, 0.5%x%Af+Cf),
+                                         cbind(0.5%x%Af+Cf, Af+Cf+Ef)), name="expCovDZf" )
> 
> ### Combine Groups
> objMZm <- mxExpectationNormal( covariance="expCovMZm", means="expMeanm", dimnames=selVars )
> objDZm <- mxExpectationNormal( covariance="expCovDZm", means="expMeanm", dimnames=selVars )
> 
> objMZf <- mxExpectationNormal( covariance="expCovMZf", means="expMeanf", dimnames=selVars )
> objDZf <- mxExpectationNormal( covariance="expCovDZf", means="expMeanf", dimnames=selVars )
> 
> parsm <- list( Xm, Ym, Zm, Alm, Clm, Elm, CovarLPm, VarLPm, pathFlm, pathAsm, pathCsm, pathEsm, covAm, covCm, covEm, covPm, matIm, invSDm, FacMeansm, FacMeans2m, meansm, expMeanm, regCoefYCm, YCCovm, regCoefOCm, OCCovm, Sflm, Unitm, averageFLm )
> parsf <- list( Xf, Yf, Zf, Alf, Clf, Elf, CovarLPf, VarLPf, pathFlf, pathAsf, pathCsf, pathEsf, covAf, covCf, covEf, covPf, matIf, invSDf, FacMeansf, FacMeans2f,  meansf, expMeanf, regCoefYCf, YCCovf, regCoefOCf, OCCovf, Sflf, Unitf, averageFLf )
> 
> 
> funML <- mxFitFunctionML()
> 
> mtfsModelMZm <- mxModel( parsm, covMZm, mtfsDataMZm, objMZm, funML, name="MZm", mxCI(c("Alm", "Clm", "Elm")))
> mtfsModelDZm <- mxModel( parsm, covDZm, mtfsDataDZm, objDZm, funML, name="DZm" )
> 
> mtfsModelMZf <- mxModel( parsf, covMZf, mtfsDataMZf, objMZf, funML, name="MZf", mxCI(c("Alf", "Clf", "Elf")))
> mtfsModelDZf <- mxModel( parsf, covDZf, mtfsDataDZf, objDZf, funML, name="DZf" )
> 
> #run MN model
> fitML <- mxFitFunctionMultigroup(c("MZm.fitfunction","DZm.fitfunction","MZf.fitfunction","DZf.fitfunction"))
> Cf <- mxModel( "ComACE", parsm, parsf, varLP1m, varLP1f, mtfsDataMZm, mtfsDataDZm, mtfsDataMZf, mtfsDataDZf, mtfsModelMZm, mtfsModelDZm, mtfsModelMZf, mtfsModelDZf, fitML)
> 
> Cf <- mxOption(Cf, "Standard Errors", "No")
> Cf <- mxOption(Cf, "Calculate Hessian", "No")
> #Cf <- mxOption(Cf, key="Number of Threads", value=4)
> CfFit <- mxRun(Cf, intervals=F) ### Run CholACE model
Running ComACE with 48 parameters
> #examine MN Output                                                           
> CfFit$output$status$code
[1] 0
> 
> modelsummary <- summary(CfFit)
> 
> modelsummary
Summary of ComACE 
 
free parameters:
       name     matrix row col      Estimate
1      x11m         Xm   1   1  4.649823e-01
2      y11m         Ym   1   1  8.340517e-01
3      z11m         Zm   1   1  2.968994e-01
4     flAFm    pathFlm   1   1  1.463656e+00
5     flAQm    pathFlm   2   1  9.932471e-01
6    flCPDm    pathFlm   3   1  1.138146e+00
7     flMJm    pathFlm   4   1  6.920370e-01
8     AFasm    pathAsm   1   1  1.237456e-01
9    ALCasm    pathAsm   1   2 -1.552780e-01
10    AQasm    pathAsm   2   2  1.949344e-01
11   CPDasm    pathAsm   3   3  1.125433e+00
12    MJasm    pathAsm   4   4  1.513553e+00
13    AFcsm    pathCsm   1   1  7.482918e-01
14   ALCcsm    pathCsm   1   2  1.206520e-01
15    AQcsm    pathCsm   2   2  4.123054e-01
16   CPDcsm    pathCsm   3   3 -1.275181e-04
17    MJcsm    pathCsm   4   4  1.190666e-05
18    AFesm    pathEsm   1   1  6.167123e-01
19   ALCesm    pathEsm   1   2 -7.291718e-02
20    AQesm    pathEsm   2   2  6.541969e-01
21   CPDesm    pathEsm   3   3  1.093856e+00
22    MJesm    pathEsm   4   4  1.304408e+00
23 beta1YCm regCoefYCm   1   1  1.795511e+00
24 beta1OCm regCoefOCm   1   1  1.714459e+00
25     x11f         Xf   1   1  4.097506e-01
26     y11f         Yf   1   1  8.119046e-01
27     z11f         Zf   1   1  4.158310e-01
28    flAFf    pathFlf   1   1  1.259678e+00
29    flAQf    pathFlf   2   1  7.662991e-01
30   flCPDf    pathFlf   3   1  7.064379e-01
31    flMJf    pathFlf   4   1  3.594333e-01
32    AFasf    pathAsf   1   1 -1.292957e-02
33   ALCasf    pathAsf   1   2  4.870555e-02
34    AQasf    pathAsf   2   2 -1.839407e-01
35   CPDasf    pathAsf   3   3  1.104443e+00
36    MJasf    pathAsf   4   4  1.030952e+00
37    AFcsf    pathCsf   1   1  7.223159e-01
38   ALCcsf    pathCsf   1   2  1.365026e-01
39    AQcsf    pathCsf   2   2  2.579595e-02
40   CPDcsf    pathCsf   3   3  2.953285e-04
41    MJcsf    pathCsf   4   4  3.630443e-01
42    AFesf    pathEsf   1   1  5.008561e-01
43   ALCesf    pathEsf   1   2 -7.096488e-02
44    AQesf    pathEsf   2   2  4.854012e-01
45   CPDesf    pathEsf   3   3  9.269777e-01
46    MJesf    pathEsf   4   4  9.343253e-01
47 beta1YCf regCoefYCf   1   1  1.729570e+00
48 beta1OCf regCoefOCf   1   1  1.618791e+00

Model Statistics: 
               |  Parameters  |  Degrees of Freedom  |  Fit (-2lnL units)
       Model:             48                  12683              41675.13
   Saturated:             NA                     NA                    NA
Independence:             NA                     NA                    NA
Number of observations/statistics: 2238/12731

Constraint 'varLP1m' contributes 1 observed statistic. 
Constraint 'varLP1f' contributes 1 observed statistic. 

Information Criteria: 
      |  df Penalty  |  Parameters Penalty  |  Sample-Size Adjusted
AIC:       16309.13               41771.13                 41773.28
BIC:      -56153.14               42045.37                 41892.87
CFI: NA 
TLI: 1   (also known as NNFI) 
RMSEA:  0  [95% CI (NA, NA)]
Prob(RMSEA <= 0.05): NA
To get additional fit indices, see help(mxRefModels)
timestamp: 2020-04-02 16:19:12 
Wall clock time: 8.621931 secs 
optimizer:  SLSQP 
OpenMx version number: 2.13.2 
Need help?  See help(mxSummary) 

> 
> round(CfFit$output$algebras$MZm.CovarLPm)
     [,1]
[1,]    1
> round(CfFit$output$algebras$MZf.CovarLPf)
     [,1]
[1,]    1
> 
> round(CfFit$output$algebras$MZm.Alm, 2)
     [,1]
[1,] 0.22
> round(CfFit$output$algebras$MZm.Clm, 2)
     [,1]
[1,]  0.7
> round(CfFit$output$algebras$MZm.Elm, 2)
     [,1]
[1,] 0.09
> 
> round(CfFit$output$algebras$MZf.Alf, 2)
     [,1]
[1,] 0.17
> round(CfFit$output$algebras$MZf.Clf, 2)
     [,1]
[1,] 0.66
> round(CfFit$output$algebras$MZf.Elf, 2)
     [,1]
[1,] 0.17
> 
> round(CfFit$output$matrices$MZm.pathAsm, 2)
      [,1]  [,2] [,3] [,4]
[1,]  0.12 -0.16 0.00 0.00
[2,] -0.16  0.19 0.00 0.00
[3,]  0.00  0.00 1.13 0.00
[4,]  0.00  0.00 0.00 1.51
> round(CfFit$output$matrices$MZm.pathCsm, 2)
     [,1] [,2] [,3] [,4]
[1,] 0.75 0.12    0    0
[2,] 0.12 0.41    0    0
[3,] 0.00 0.00    0    0
[4,] 0.00 0.00    0    0
> round(CfFit$output$matrices$MZm.pathEsm, 2)
      [,1]  [,2] [,3] [,4]
[1,]  0.62 -0.07 0.00  0.0
[2,] -0.07  0.65 0.00  0.0
[3,]  0.00  0.00 1.09  0.0
[4,]  0.00  0.00 0.00  1.3
> 
> round(CfFit$output$matrices$MZf.pathAsf, 2)
      [,1]  [,2] [,3] [,4]
[1,] -0.01  0.05  0.0 0.00
[2,]  0.05 -0.18  0.0 0.00
[3,]  0.00  0.00  1.1 0.00
[4,]  0.00  0.00  0.0 1.03
> round(CfFit$output$matrices$MZf.pathCsf, 2)
     [,1] [,2] [,3] [,4]
[1,] 0.72 0.14    0 0.00
[2,] 0.14 0.03    0 0.00
[3,] 0.00 0.00    0 0.00
[4,] 0.00 0.00    0 0.36
> round(CfFit$output$matrices$MZf.pathEsf, 2)
      [,1]  [,2] [,3] [,4]
[1,]  0.50 -0.07 0.00 0.00
[2,] -0.07  0.49 0.00 0.00
[3,]  0.00  0.00 0.93 0.00
[4,]  0.00  0.00 0.00 0.93
> 
> round(CfFit$output$matrices$MZm.pathFlm, 2)
     [,1]
[1,] 1.46
[2,] 0.99
[3,] 1.14
[4,] 0.69
> round(CfFit$output$matrices$MZf.pathFlf, 2)
     [,1]
[1,] 1.26
[2,] 0.77
[3,] 0.71
[4,] 0.36
> 
> round(CfFit$output$algebras$MZm.Sflm, 2)
     [,1]
[1,] 0.83
[2,] 0.77
[3,] 0.59
[4,] 0.33
> round(CfFit$output$algebras$MZf.Sflf, 2)
     [,1]
[1,] 0.82
[2,] 0.82
[3,] 0.44
[4,] 0.24
> 
> round(CfFit$output$algebras$MZm.averageFLm, 2)
     [,1]
[1,] 0.63
> round(CfFit$output$algebras$MZf.averageFLf, 2)
     [,1]
[1,] 0.58
> 
> round(CfFit$output$matrices$MZm.FacMeansm, 2)
     [,1]
[1,]    0
> round(CfFit$output$matrices$MZf.FacMeansf, 2)
     [,1]
[1,]    0
> 
> round(CfFit$output$algebras$MZm.FacMeans2m, 2)
     [,1]
[1,] 1.71
> round(CfFit$output$algebras$MZf.FacMeans2f, 2)
     [,1]
[1,] 1.62
> 
> round(CfFit$output$algebras$MZm.expMeanm, 2)
     [,1] [,2] [,3] [,4] [,5] [,6] [,7] [,8]
[1,] 2.51  1.7 1.95 1.19 2.51  1.7 1.95 1.19
> round(CfFit$output$algebras$MZf.expMeanf, 2)
     [,1] [,2] [,3] [,4] [,5] [,6] [,7] [,8]
[1,] 2.04 1.24 1.14 0.58 2.04 1.24 1.14 0.58
> 
> 
> round(CfFit$output$matrices$MZm.regCoefOCm, 2)
     [,1]
[1,] 1.71
> round(CfFit$output$matrices$MZm.regCoefYCm, 2)
     [,1]
[1,]  1.8
> 
> round(CfFit$output$matrices$MZf.regCoefOCf, 2)
     [,1]
[1,] 1.62
> round(CfFit$output$matrices$MZf.regCoefYCf, 2)
     [,1]
[1,] 1.73
>