dprior: Prior Density

In abtest: Bayesian A/B Testing

Description

Function for evaluating the prior density.

Usage

dprior(
  x1,
  x2 = NULL,
  prior_par = list(mu_psi = 0, sigma_psi = 1, mu_beta = 0, sigma_beta = 1),
  what = "logor",
  hypothesis = "H1"
)

Arguments

x1	numeric vector with values at which the prior density should be evaluated.
x2	if what = "p1p2", value of p2 (i.e., the latent "success" probability in the experimental condition) at which the joint prior density should be evaluated. If what = "p2givenp1", the given value of p1 (i.e., the latent "success" probability in the control condition).
prior_par	list with prior parameters. This list needs to contain the following elements: mu_psi (prior mean for the normal prior on the test-relevant log odds ratio), sigma_psi (prior standard deviation for the normal prior on the test-relevant log odds ratio), mu_beta (prior mean for the normal prior on the grand mean of the log odds), sigma_beta (prior standard deviation for the normal prior on the grand mean of the log odds). Each of the elements needs to be a real number (the standard deviations need to be positive). The default are standard normal priors for both the log odds ratio parameter and the grand mean of the log odds parameter.
what	character specifying for which quantity the prior density should be evaluated. Either "logor" (i.e., log odds ratio) , "or" (i.e., odds ratio), "p1p2" (i.e., the joint density of the latent "success" probability in the experimental and control condition), "p1" (i.e., latent "success" probability in the control condition), "p2" (i.e., latent "success" probability in the experimental condition), "p2givenp1" (i.e., conditional distribution of the latent "success" probability in the experimental condition given a "success" probability of p1 in the control condition), "rrisk" (i.e., relative risk, the ratio of the "success" probability in the experimental and the control condition), or "arisk" (i.e., absolute risk, the difference of the "success" probability in the experimental and control condition).
hypothesis	character specifying whether to evaluate the two-sided prior density (i.e., "H1"), the one-sided prior density with lower truncation point (i.e., "H+"), or the one-sided prior density with upper truncation point (i.e., "H-").

Value

numeric vector with the values of the prior density.

Note

Internally, the test-relevant prior is always a normal prior on the log odds ratio, consequently, if what is not "logor", the implied prior density for the quantity is returned.

Author(s)

Quentin F. Gronau

Examples

# prior parameters
prior_par <- list(mu_psi = 0, sigma_psi = 1,
                  mu_beta = 0, sigma_beta = 1)

# prior density
dprior(x1 = 0.1, prior_par = prior_par, what = "logor")
dprior(x1 = 1.1, prior_par = prior_par, what = "or")
dprior(x1 = 0.49, x2 = 0.51, prior_par = prior_par, what = "p1p2")
dprior(x1 = 0.45, prior_par = prior_par, what = "p1")
dprior(x1 = 0.45, prior_par = prior_par, what = "p2")
dprior(x1 = 0.49, x2 = 0.51, prior_par = prior_par, what = "p2givenp1")
dprior(x1 = 1.05, prior_par = prior_par, what = "rrisk")
dprior(x1 = 0.02, prior_par = prior_par, what = "arisk")

# also works for vectors
dprior(x1 = c(-0.1, 0, 0.1, 0.2), prior_par = prior_par, what = "logor")

abtest documentation built on Nov. 22, 2021, 9:07 a.m.