tests/testthat/test_mllogis.R

In univariateML: Maximum Likelihood Estimation for Univariate Densities

context("mllogis")

## Data generation.
set.seed(313)
tiny_data <- stats::rlogis(10, 1, 7)
small_data <- stats::rlogis(100, 10, 3)
medium_data <- stats::rlogis(1000, 1 / 2, 2)
large_data <- stats::rlogis(10000, 20, 13)

## Checks if the ML is correct.

m <- stats::median(tiny_data)
mad <- stats::median(abs(tiny_data - m))
start <- c(m, log(mad))

mle1 <- suppressWarnings(nlm(function(p) {
  -sum(stats::dlogis(tiny_data, p[1], exp(p[2]), log = TRUE))
}, p = start))


m <- stats::median(small_data)
mad <- stats::median(abs(small_data - m))
start <- c(m, log(mad))

mle2 <- suppressWarnings(nlm(function(p) {
  -sum(stats::dlogis(small_data, p[1], exp(p[2]), log = TRUE))
}, p = start))


m <- stats::median(medium_data)
mad <- stats::median(abs(medium_data - m))
start <- c(m, log(mad))

mle3 <- suppressWarnings(nlm(function(p) {
  -sum(stats::dlogis(medium_data, p[1], exp(p[2]), log = TRUE))
}, p = start))

m <- stats::median(large_data)
mad <- stats::median(abs(large_data - m))
start <- c(m, log(mad))

mle4 <- suppressWarnings(nlm(function(p) {
  -sum(stats::dlogis(large_data, p[1], exp(p[2]), log = TRUE))
}, p = start))

## Checks estimates.
expect_equal(c(mle1$estimate[1], exp(mle1$estimate[2])),
  as.numeric(mllogis(tiny_data)),
  tolerance = 1e-5
)
expect_equal(c(mle2$estimate[1], exp(mle2$estimate[2])),
  as.numeric(mllogis(small_data)),
  tolerance = 1e-5
)
expect_equal(c(mle3$estimate[1], exp(mle3$estimate[2])),
  as.numeric(mllogis(medium_data)),
  tolerance = 1e-5
)
expect_equal(c(mle4$estimate[1], exp(mle4$estimate[2])),
  as.numeric(mllogis(large_data)),
  tolerance = 1e-5
)

## Checks logLiks.
expect_equal(-mle1$minimum, attr(mllogis(tiny_data), "logLik"),
  tolerance = 1e-5
)
expect_equal(-mle2$minimum, attr(mllogis(small_data), "logLik"),
  tolerance = 1e-5
)
expect_equal(-mle3$minimum, attr(mllogis(medium_data), "logLik"),
  tolerance = 1e-5
)
expect_equal(-mle4$minimum, attr(mllogis(large_data), "logLik"),
  tolerance = 1e-5
)

## Finds errors with na and data out of bounds.
expect_error(mllogis(c(tiny_data, NA)))

## Checks that na.rm works as intended.
expect_equal(
  coef(mllogis(small_data)),
  coef(mllogis(c(small_data, NA), na.rm = TRUE))
)

## Check class.
est <- mllogis(small_data, na.rm = TRUE)
expect_equal(attr(est, "model"), "Logistic")
expect_equal(class(est), "univariateML")