Unnecessary checks in check_model for NB models

[mattansb]

Dispersion plot does not seem to account for the theta parameter in the model?

library(performance)

set.seed(3)
mu <- rpois(500, lambda = 3)
x <- rnorm(500, mu, mu*3) |> ceiling() |> pmax(0)

quine.nb1 <- MASS::glm.nb(x ~ mu)


check_model(quine.nb1)

^{Created on 2022-10-23 by the reprex package (v2.0.1)}

[strengejacke]

@bwiernik ?

[strengejacke]

Not sure, do you still think the plot is inaccurate? The results from simulate_residuals() (DHARMa) look in line with the plot?

library(performance)

set.seed(3)
mu <- rpois(500, lambda = 3)
x <- rnorm(500, mu, mu * 3) |>
  ceiling() |>
  pmax(0)
m <- MASS::glm.nb(x ~ mu)

check_overdispersion(simulate_residuals(m))
#> # Overdispersion test
#> 
#>  dispersion ratio =   0.410
#>           p-value = < 0.001
#> Underdispersion detected.

check_model(m)

^{Created on 2024-03-17 with reprex v2.1.0}

[strengejacke]

(edit: sorry, wrong reprex, did not reinstall the pkg)