Description Usage Arguments Details Author(s) References See Also Examples
Calculates the posterior distribution given data data
and prior priormix
, where the prior is an empirical Bayes powerprior (Gravestock&Held, 2017). Supports beta-binomial and normal-normal model.
1 2 3 4 5 6 | ## S3 method for class 'powerprior'
postmix(priormix, data, n, r, m, se, p.prior.a, p.prior.b, ...)
## S3 method for class 'betaMix'
powerprior(prior, n, r, p.prior.a, p.prior.b, ...)
## S3 method for class 'normMix'
powerprior(prior, n, m, sigma, ...)
|
priormix |
prior object created by |
prior |
An RBesT mixture object with a single mixture component |
data |
individual data as in |
n |
sample size |
r |
number of successes |
m |
sample mean |
se |
sample standard error |
sigma |
standard deviation |
p.prior.a |
in case of binary outcome, shape1 parameter of initial beta prior for successes |
p.prior.b |
in case of binary outcome, shape2 parameter of initial beta prior for successes |
... |
currently not supported |
Extends postmix
. See its documentation for further details on arguments.
powerprior
and postmix.powerprior
are equivalent except that the former accepts a single component mixture object while the latter accepts a powerprior
object created by as.powerprior
. powerprior
may be deprecated in the future.
Manuel Wiesenfarth
Gravestock, I. and Held, L. (2017). Adaptive power priors with empirical bayes for clinical trials. Pharmaceutical statistics, 16(5):349-360.
package StudyPrior
1 2 3 4 5 6 7 8 9 10 11 12 | ######################
# Normal Outcome
# standard deviation
sigma=1
# prior with nominal prior ESS=50
inf=c(0,1/sqrt(50))
info <-mixnorm(informative=c(1, inf), sigma=sigma)
n=10
posterior=postmix(as.powerprior(info),n = n,m=1,se=sigma/sqrt(n))
plot(posterior)
pmix(posterior, 1.5, lower.tail = FALSE)
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.