plot_robustness: Plot Bayes Factor Robustness Check

In quentingronau/abtest: Bayesian A/B Testing

Description

Function for plotting Bayes factor robustness check results (i.e., prior sensitivity analysis).

Usage

plot_robustness(
  x,
  bftype = "BF10",
  log = FALSE,
  mu_range = c(0, 0.3),
  sigma_range = c(0.25, 1),
  mu_steps = 40,
  sigma_steps = 40,
  cores = 1,
  ...
)

Arguments

x	object of class "ab".
bftype	character that specifies which Bayes factor is plotted. Either "BF10", "BF01", "BF+0", "BF0+", "BF-0", or "BF0-".
log	Boolean that specifies whether the log Bayes factor is plotted.
mu_range	numeric vector of length two that specifies the range of mu_psi values to consider.
sigma_range	numeric vector of length two that specifies the range of sigma_psi values to consider.
mu_steps	numeric value that specifies in how many discrete steps the interval mu_range is partitioned.
sigma_steps	numeric value that specifies in how many discrete steps the interval sigma_range is partitioned.
cores	number of cores used for the computations.
...	further arguments passed to filled.contour.

Details

The plot shows how the Bayes factor changes as a function of the normal prior location parameter mu_psi and the normal prior scale parameter sigma_psi (i.e., a prior sensitivity analysis with respect to the normal prior on the test-relevant log odds ratio).

Value

Returns a data.frame with the mu_psi values, sigma_psi values, and corresponding (log) Bayes factors.

Author(s)

Quentin F. Gronau

Examples

## Not run: 
# synthetic data
data <- list(y1 = 10, n1 = 28, y2 = 14, n2 = 26)

# Bayesian A/B test with default settings
ab <- ab_test(data = data)

# plot robustness check (i.e., prior sensitivity analysis)
p <- plot_robustness(ab)

# returned object contains the Bayes factors for the different prior settings
head(p)

## End(Not run)

quentingronau/abtest documentation built on Nov. 23, 2021, 1:43 a.m.