tests/testthat/test_pearson.R

In msm: Multi-State Markov and Hidden Markov Models in Continuous Time

test_that("pearson.msm help example",{
  psor.q <- rbind(c(0,0.1,0,0),c(0,0,0.1,0),c(0,0,0,0.1),c(0,0,0,0))
  psor.msm <- msm(state ~ months, subject=ptnum, data=psor,
                  qmatrix = psor.q, covariates = ~ollwsdrt+hieffusn,
                  constraint = list(hieffusn=c(1,1,1),ollwsdrt=c(1,1,2)))
  p <- pearson.msm(psor.msm, timegroups=2, intervalgroups=2, covgroups=2)
  expect_equal(p$test[["p"]],0)

  # with exact death times
  set.seed(1)
  cav.msm <- msm( state ~ years, subject=PTNUM, data = cav[1:1000,],
                  qmatrix = twoway4.q, deathexact = TRUE, fixedpars=FALSE)
  p <- pearson.msm(cav.msm)
  expect_equal(p$test[["stat"]],67,tol=1)
})

#test_that("pearson with bootstrap",{
#  psor.msm <- msm(state ~ months, subject=ptnum, data=psor[1:29,],
#                  qmatrix = psor.q)
#  p <- pearson.msm(psor.msm, timegroups=2, intervalgroups=2, covgroups=1, boot=TRUE, B=10)
                                        #})

test_that("pearson.msm, models with interactions",{
  skip_on_cran()
  expect_error({
    psor.q <- rbind(c(0,0.1,0,0),c(0,0,0.1,0),c(0,0,0,0.1),c(0,0,0,0))
    psor.msm <- msm(state ~ months, subject=ptnum, data=psor,
                    qmatrix = psor.q, covariates = ~ollwsdrt:hieffusn)
    pearson.msm(psor.msm, timegroups=2, intervalgroups=2, covgroups=2)
  }, NA)
})