test_uniformity: Test uniformity of PIT residuals

In tulpa: Templated Unified Library for Posterior Approximation in Bayesian Hierarchical Models

Test uniformity of PIT residuals

Description

If the model is correct, PIT residuals should follow Uniform(0, 1). Applies a Kolmogorov-Smirnov test against that null. Equivalent to DHARMa::testUniformity().

Usage

test_uniformity(
  object,
  observed = NULL,
  nsim = 250L,
  seed = 123L,
  plot = FALSE
)

Arguments

object	A fitted model, a numeric vector of PIT residuals, or a matrix of simulations
observed	Observed data (if object is simulations matrix)
nsim	Number of simulations (default 250)
seed	Random seed (default 123)
plot	If TRUE, draws a QQ plot

Value

An htest object (KS test result)

Examples

set.seed(1)
test_uniformity(runif(50))   # PIT residuals supplied directly

set.seed(1)
d <- data.frame(x = rnorm(80))
d$y <- rpois(80, exp(0.3 + 0.5 * d$x))
fit <- tulpa(y ~ x, data = d, family = "poisson", mode = "laplace")
test_uniformity(fit, nsim = 50)

tulpa documentation built on Oct. 19, 2026, 5:09 p.m.