pprior {abtest} | R Documentation |
Function for evaluating the prior cumulative distribution function (CDF).
pprior( q, prior_par = list(mu_psi = 0, sigma_psi = 1, mu_beta = 0, sigma_beta = 1), what = "logor", hypothesis = "H1" )
q |
numeric vector with quantiles. |
prior_par |
list with prior parameters. This list needs to contain the
following elements: |
what |
character specifying for which quantity the prior CDF should be
evaluated. Either |
hypothesis |
character specifying whether to evaluate the CDF for a two-sided prior (i.e., "H1"), a one-sided prior with lower truncation point (i.e., "H+"), or a one-sided prior with upper truncation point (i.e., "H-"). |
numeric vector with the values of the prior CDF.
Internally, the test-relevant prior is always a normal prior on the log
odds ratio, consequently, if what
is not "logor"
, the
implied prior CDF for the quantity is returned.
Quentin F. Gronau
# prior parameters prior_par <- list(mu_psi = 0, sigma_psi = 1, mu_beta = 0, sigma_beta = 1) # evaluate prior CDF pprior(q = 0.1, prior_par = prior_par, what = "logor") pprior(q = 1.1, prior_par = prior_par, what = "or") pprior(q = 1.05, prior_par = prior_par, what = "rrisk") pprior(q = 0.02, prior_par = prior_par, what = "arisk") # also works for vectors pprior(q = c(-0.1, 0, 0.1, 0.2), prior_par = prior_par, what = "logor")