tests/testthat/test-sigma2_cnb.R

In cnbdistr: Conditional Negative Binomial Distribution

context("sigma2_cnb")

test_that("sigma2_cnb stops when it should", {

  expect_error( sigma2_cnb(120, 90, 0.994) ) # missing an argument
  expect_error( sigma2_cnb(-2000, 120, 90, 0.994) ) # D < 0
  expect_error( sigma2_cnb(2000, -120, 90, 0.994) ) # r1 < 0
  expect_error( sigma2_cnb(2000, 120, -90, 0.994) ) # r2 < 0
  expect_error( sigma2_cnb(2000, 120, 90, -0.994) ) # lambda < 0

})

test_that("sigma2_cnb computes correctly", {

  # lamba == 1
  D <- 32
  r1 <- 600
  r2 <- 499
  lambda <- 1
  expect_equal(sigma2_cnb(D, r1, r2, lambda),  sum(rmutil::dbetabinom(0:D, D, r1/(r1+r2), r1+r2) * (0:D - mu_cnb(D, r1, r2, lambda))^2))

  # lambda < 1
  D <- 64
  r1 <- 563
  r2 <- 49
  lambda <- 0.72
  expect_equal(sigma2_cnb(D, r1, r2, lambda),  sum(dcnb(0:D, D, r1, r2, lambda) * (0:D - mu_cnb(D, r1, r2, lambda))^2))

  # lambda > 1
  D <- 99
  r1 <- 60
  r2 <- 588
  lambda <- 1.32
  expect_equal(sigma2_cnb(D, r1, r2, lambda),  sum(dcnb(0:D, D, r1, r2, lambda) * (0:D - mu_cnb(D, r1, r2, lambda))^2))


})