Unit tests of msm are not passing fully

[Andreas Tille]

Hi Christopher,

I try to upgrade the Debian package of msm to the latest version but
failed tu run the according test suite:


> library(testthat)
> test_check("msm")
Loading required package: msm
analytic transition probability matrices : ...................................................
msm data summaries : ...1..2........
analytic derivatives of likelihood : ............
analytic derivatives of likelihood in HMMs : ...................
Hidden Markov model likelihoods : .......
Hidden Markov model error handling : 34.........5.....
msm misclassification model likelihoods : ....................
msm simple model likelihoods : .........................
censored states : .......................
output functions : ...................................................................................................6..................78......9a......b.c................................................................
Distribution functions and utilities : ...................................................


1. Failure(@test_datasumm.r#8): statetable.msm ---------------------------------
statetable.msm(state, PTNUM) does not match 'not found'. Actual value: "Error in statetable.msm(state, PTNUM) : Objekt 'state' nicht gefunden\n"

2. Failure(@test_datasumm.r#16): crudeinits.msm --------------------------------
crudeinits.msm(state ~ years, PTNUM, qmatrix = twoway4.q) does not match 'not found'. Actual value: "Error in eval(expr, envir, enclos) : Objekt 'state' nicht gefunden\n"

3. Failure(@test_models_hmm.r#57): errors in hmodel ----------------------------
hmodel3 <- list(hmmMETNorm(mean = 100, sd = 16, splat = 8, lower = 80, upper = Inf, meanerr = 0), hmmMETNorm(mean = 54, sd = 18, sderr = 8, lower = 0, upper = 80, meanerr = 0), hmmIdent(999)) does not match 'unused argument \(splat'. Actual value: "Error in hmmMETNorm(mean = 100, sd = 16, splat = 8, lower = 80, upper = Inf,  : \n  unbenutztes Argument (splat = 8)\n"

4. Failure(@test_models_hmm.r#63): errors in hmodel ----------------------------
hmodel3 <- list(hmmMETNorm(mean = 100, sd = 16, sderr = 8, lower = 80, upper = Inf, meanerr = 0), hmmMETNorm(mean = 54, sd = 18, sderr = 8, lower = 0, upper = 80, meanerr = 0), hmmIdent(999, 3)) does not match 'unused argument \(3'. Actual value: "Error in hmmIdent(999, 3) : unbenutztes Argument (3)\n"

5. Failure(@test_models_hmm.r#127): error handling in HMM fits -----------------
fev3.hid <- msm(fev ~ days, subject = ptnum, data = fev, qmatrix = three.q, death = 3, hmodel = hmodel3, hcovariates = list(~acute, ~nonexistent, NULL), hcovinits = list(-8, -8, NULL)) does not match 'object .+ not found'. Actual value: "Error in eval(expr, envir, enclos) : Objekt 'nonexistent' nicht gefunden\n"

6. Failure(@test_models.r#422): qmatrix subset function ------------------------
Q[1, 2, 3, 4, 5] does not match 'unused arguments'. Actual value: "Error in `[.msm.est`(Q, 1, 2, 3, 4, 5) : unbenutzte Argumente (4, 5)\n"

7. Failure(@test_models.r#498): error handling: formula ------------------------
cav.msm <- msm(state, subject = PTNUM, data = cav, qmatrix = twoway4.q, death = TRUE, fixedpars = TRUE) does not match 'not found'. Actual value: "Error in inherits(formula, "formula") : Objekt 'state' nicht gefunden\n"

8. Failure(@test_models.r#499): error handling: formula ------------------------
cav.msm <- msm(~1, subject = PTNUM, data = cav, qmatrix = twoway4.q, death = TRUE, fixedpars = TRUE) does not match 'invalid data'. Actual value: "Error in as.name(all.vars(formula[[2]])) : \n  ungültige Daten vom Mode 'character' (zu kurz)\n"

9. Failure(@test_models.r#516): error handling: subject ------------------------
cav.msm <- msm(state ~ years, subject = "foo", data = cav, qmatrix = twoway4.q, death = TRUE, fixedpars = TRUE) does not match 'variable lengths differ'. Actual value: "Error in model.frame.default(data = structure(list(PTNUM = c(100002L,  : \n  Variablenlängen sind unterschiedlich (gefunden für '(state)')\n"

a. Failure(@test_models.r#517): error handling: subject ------------------------
cav.msm <- msm(state ~ years, subject = foo, data = cav, qmatrix = twoway4.q, death = TRUE, fixedpars = TRUE) does not match 'not found'. Actual value: "Error in eval(expr, envir, enclos) : Objekt 'foo' nicht gefunden\n"

b. Failure(@test_models.r#533): error handling: covariates ---------------------
cav.msm <- msm(state ~ years, subject = PTNUM, data = cav, qmatrix = twoway4.q, covariates = ~sux) does not match 'not found'. Actual value: "Error in eval(expr, envir, enclos) : Objekt 'sux' nicht gefunden\n"

c. Failure(@test_models.r#535): error handling: covariates ---------------------
cav.msm <- msm(state ~ years, subject = PTNUM, data = cav, qmatrix = twoway4.q, covariates = ~sox) does not match 'not found'. Actual value: "Error in eval(expr, envir, enclos) : Objekt 'sox' nicht gefunden\n"

Fehler: Test failures
Ausführung angehalten



Do you need any further information to track this down?

Kind regards and thanks for providing msm as free software

       Andreas.

-- 
http://fam-tille.de