ROPE: Conduct a ROPE test

In abnormally-distributed/Bayezilla: Bayesian Models With JAGS and Several Utilities for Data Exploration and Model Evaluation.

Description

Conduct a test of practical equivalence by defining a region of practical equivalence (ROPE) around the null hypothesis H0. This is most appropriate when you wish to test whether an effect is larger in magnitude than a cutoff that defines an effect size that is practically unimportant. Hence, this is a test for practical significance and requires a well defined idea of what practical is.

Usage

ROPE(fit, ROPE = c(-0.1, 0.1), H0 = 0, plot = TRUE, param = NULL,
  line.color = "black", interval.color = "#8c00ffCC",
  density.color = "#2CBAFFCC", rope.color = "#FF0000CC",
  interval.method = "QI", cred.level = 0.9)

Arguments

fit	a stanfit or runjags object.
ROPE	a numeric vector of two numbers. Defaults to c(-.1, .1), assuming a Cohen's d... Adjust to what is reasonable!
plot	plot the output. defaults to TRUE.
param	the name of the parameter in the stanfit or runjags object you want to plot.
line.color	color marking H0. "black" is default.
interval.color	defaults to purple/violet "#8c00ffCC"
density.color	defaults to blue "#2CBAFFCC"
rope.color	defaults to red "#FF0000CC"
interval.method	One of "QI" (the default), "HDI"
cred.level	the confidence level for interval.method. Defaults to .90

Examples

ROPE()

abnormally-distributed/Bayezilla documentation built on Oct. 31, 2019, 1:57 a.m.