tests/testthat/test-knownFits.R

In ctsemOMX: Continuous Time SEM - 'OpenMx' Based Functions

if(identical(Sys.getenv("NOT_CRAN"), "true")& .Machine$sizeof.pointer != 4){
library(ctsemOMX)
library(testthat)

context("knownFits")

#anomauth
test_that("anomauth", {

  data(AnomAuth)
  AnomAuthmodel<-ctModel(LAMBDA=matrix(c(1, 0, 0, 1), nrow=2, ncol=2),  
    n.latent=2,n.manifest=2, 
    MANIFESTVAR=diag(0,2),
    Tpoints=5)

  ll=Inf
  counter=0
  while(counter < 10 && (ll >  23415.99  || ll < 23415.0)){
    counter=counter+1
  AnomAuthfit<-ctFit(AnomAuth, AnomAuthmodel)
  ll=AnomAuthfit$mxobj$output$Minus2LogLikelihood
  }
  
  expect_equal(23415.929,ll)


})

# #anomauth with trait asymptotic vs standard param comparisons ##started failing due to optimizer instability
# test_that("AnomAuth_traitasymptoticcheck", {
#   
#   data(AnomAuth)
#   AnomAuthmodel<-ctModel(LAMBDA=matrix(c(1, 0, 0, 1), nrow=2, ncol=2),  
#     n.latent=2,n.manifest=2, 
#     MANIFESTVAR=diag(0,2),
#     TRAITVAR='auto',
#     CINT=matrix(0,nrow=2),
#     T0MEANS=matrix(0,nrow=2),
#     MANIFESTMEANS=matrix(c('m1','m2'),nrow=2),
#     Tpoints=5)
#   
#   AnomAuthfit2<-ctFit(AnomAuth, AnomAuthmodel,asymptotes=TRUE, verbose=0,retryattempts=3)
#   AnomAuthfit1<-ctFit(AnomAuth, AnomAuthmodel,asymptotes=FALSE, verbose=0,retryattempts=3)
#   
#   
#   expect_equal(AnomAuthfit2$mxobj$output$Minus2LogLikelihood,AnomAuthfit1$mxobj$output$Minus2LogLikelihood)
#   
#   summ1<-summary(AnomAuthfit1,verbose=TRUE)
#   summ2<-summary(AnomAuthfit2,verbose=TRUE)
#   
#   expect_equal(summ1$ctparameters[,1:2],summ2$ctparameters[,1:2],tolerance = .001)
#   
# })

# 
# test_that("oscillator", {
# data("Oscillating")
# 
# inits <- c(-39.5, -.5, .1, 1, 0, 1, 0.05, .9)
# names(inits) <- c("crosseffect","autoeffect", "diffusion",
#   "T0var11", "T0var21", "T0var22","m1", "m2")
# 
# oscillatingm <- ctModel(n.latent = 2, n.manifest = 1, Tpoints = 11, 
#   MANIFESTVAR = matrix(c(0), nrow = 1, ncol = 1),
#   LAMBDA = matrix(c(1, 0), nrow = 1, ncol = 2),
#   T0MEANS = matrix(c('m1', 'm2'), nrow = 2, ncol = 1), 
#   T0VAR = matrix(c("T0var11", "T0var21", 0, "T0var22"), nrow = 2, ncol = 2),
#   DRIFT = matrix(c(1e-5, "crosseffect", 1, "autoeffect"), nrow = 2, ncol = 2), 
#   CINT = matrix(0, ncol = 1, nrow = 2),
#   DIFFUSION = matrix(c(0, 0, 0, "diffusion"), nrow = 2, ncol = 2))#,
#   # startValues = inits)
# 
# oscillatingf <- ctFit(Oscillating, oscillatingm, carefulFit = FALSE,retryattempts = 0)
# 
# expect_equal(-3461.936,oscillatingf$mxobj$output$Minus2LogLikelihood,tolerance=.001)
# 
# 
# })

}