tests/testthat/test_LOGISTIC.R

In NVE/fitdistrib: A list a functions to fit data to probability distributions

# Testing the GL functions
# Problem: fitting functions for GL do not return the standard error
# So the 95% confidence interval was replaced by +-20% accuracy

param <- c(45, 2.5)  # Approximate parameters taken from the fit of the Narsjo "2.11" station data
CI <- 2  # +-2 times the Std_error to get 95% confidence interval

random_distrib <- rlogis(1000, param[1],param[2])

####
test <- logist_mle(random_distrib)

 test_that("gl_mle return list of correct length", {
   expect_length(test$estimate, 2)
   expect_length(test$se, 2)
 })

 test_that("gl_mle returns reasonable estimates", {

#    expected_min <- param * (1 - CI * test$se)
#    expected_max <- param * (1 + CI * test$se)

   expected_min <- param * (1 - 0.2)
   expected_max <- param * (1 + 0.2)

   expect_true(test$estimate[1] < expected_max[1]  && test$estimate[1] > expected_min[1])
   expect_true(test$estimate[2] < expected_max[2]  && test$estimate[2] > expected_min[2])
   # There is aproblem with the 3rd parameter returned by gl_mle which is 0.1 instead of -0.3
   # Warning: Maybe a different distribution formulation?
   # expect_true(test$estimate[3] < expected_min[3]  && test$estimate[3] > expected_max[3])  # Test is the other way around because param is < 0
 })


####
test <- logist_Lmom(random_distrib)

test_that("gl_Lmom return list of correct length", {
  expect_length(test$estimate, 2)
  expect_length(test$se, 2)
})

test_that("gl_Lmom returns reasonable estimates", {

  #    expected_min <- param * (1 - CI * test$se)
  #    expected_max <- param * (1 + CI * test$se)

  expected_min <- param * (1 - 0.2)
  expected_max <- param * (1 + 0.2)

  expect_true(test$estimate[1] < expected_max[1]  && test$estimate[1] > expected_min[1])
  expect_true(test$estimate[2] < expected_max[2]  && test$estimate[2] > expected_min[2])
#  expect_true(test$estimate[3] < expected_min[3]  && test$estimate[3] > expected_max[3])  # Test is the other way around because param is < 0
})

####
test <- logist_mom(random_distrib)

test_that("gl_mom return list of correct length", {
  expect_length(test$estimate, 2)
  expect_length(test$se, 2)
})

test_that("gl_mom returns reasonable estimates", {

  #    expected_min <- param * (1 - CI * test$se)
  #    expected_max <- param * (1 + CI * test$se)

  expected_min <- param * (1 - 0.2)
  expected_max <- param * (1 + 0.2)

  expect_true(test$estimate[1] < expected_max[1]  && test$estimate[1] > expected_min[1])
  expect_true(test$estimate[2] < expected_max[2]  && test$estimate[2] > expected_min[2])
  # 3rd estimated parameter is negative but not very accurate...
  # expect_true(test$estimate[3] < expected_min[3]  && test$estimate[3] > expected_max[3])  # Test is the other way around because param is < 0
})