SCL: The SCL distribution

In sbim: Simulation-Based Inference using a Metamodel for Log-Likelihood Estimator

The SCL distribution

Description

Quantile function, distribution function, and random generation for the SCL distribution family. See Park (2025) for information about the SCL distributions.

Usage

qscl(
  p,
  M,
  k,
  num_error_size = 0.01,
  lower = TRUE,
  log_p = FALSE,
  force = FALSE
)

pscl(
  q,
  M,
  k,
  num_error_size = 0.01,
  lower = TRUE,
  log_p = FALSE,
  force = FALSE
)

rscl(n, M, k)

Arguments

p	vector of probabilities
M	the first parameter for the SCL distributions
k	the second parameter for the SCL distribution
num_error_size	The requested size of numerical error for the outputs of qscl and pscl functions, in terms of the estimated standard deviation of the output. For example num_error_size of 0.01 will output values with the standard deviation of approximately equal to 0.01.
lower	logical; if TRUE, probabilities are P(X <= x), otherwise, P(X > x).
log_p	logical; if TRUE, probabilities p are given as log(p).
force	logical; if TRUE, the function will run regardless of how long it will take. If FALSE, the function will ask if you want to continue, stop, or give a new num_error_size value whenever the expected run time is longer than 15 seconds.
q	vector of quantiles
n	number of draws

Value

a list consisting of the numeric vector of quantiles and the num_error_size (numeric) used.

sbim documentation built on April 4, 2025, 5:59 a.m.