Description Usage Arguments Details Value Methods (by class) References See Also Examples
Add a noninformative component to a mixture prior.
1 2 3 4 5 6 7 8 9 10 
priormix 
orior (mixture of conjugate distributions). 
weight 
weight given to the noninformative component (0 < 
mean 
mean of the noninformative component. It is recommended to set this parameter explicitly. 
n 
number of observations the noninformative prior corresponds to, defaults to 1. 
... 
optional arguments are ignored. 
sigma 
Sampling standard deviation for the case of Normal mixtures. 
It is recommended to robustify informative priors derived
with gMAP
using unitinformation priors . This
protects against priordata conflict, see for example
Schmidli et al., 2015.
The procedure can be used with beta, gamma and normal mixture
priors. A unitinformation prior (see Kass and Wasserman,
1995) corresponds to a prior which represents the observation of
n=1 at the null hypothesis. As the null is problem dependent we
strongly recommend to make use of the mean
argument
accordingly. See below for the definition of the default mean.
The weights of the mixture priors are rescaled to (1weight)
while the noninformative prior is assigned the weight
given.
New mixture with an extra noninformative component named
robust
.
betaMix
: The default mean
is set to 1/2 which
represents no difference between the occurrence rates for one of the
two outcomes. As the uniform Beta(1,1)
is more appropriate in
practical applications, RBesT
uses n+1
as the sample
size such that the default robust prior is the uniform instead of
the Beta(1/2,1/2)
which strictly defined would be the unit
information prior in this case.
gammaMix
: The default mean
is set to the mean of the
prior mixture. It is strongly recommended to explicitly set the
mean to the location of the null hypothesis.
normMix
: The default mean
is set to the mean
of the prior mixture. It is strongly recommended to explicitly set
the mean to the location of the null hypothesis, which is very
often equal to 0. It is also recommended to explicitly set the
sampling standard deviation using the sigma
argument.
Schmidli H, Gsteiger S, Roychoudhury S, O'Hagan A, Spiegelhalter D, Neuenschwander B. Robust metaanalyticpredictive priors in clinical trials with historical control information. Biometrics 2014;70(4):10231032.
Kass RE, Wasserman L A Reference Bayesian Test for Nested Hypotheses and its Relationship to the Schwarz Criterion J Amer Statist Assoc 1995; 90(431):928934.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17  bmix < mixbeta(inf1=c(0.2, 8, 3), inf2=c(0.8, 10, 2))
plot(bmix)
rbmix < robustify(bmix, weight=0.1, mean=0.5)
rbmix
plot(rbmix)
gmnMix < mixgamma(inf1=c(0.2, 2, 3), inf2=c(0.8, 2, 5), param="mn")
plot(gmnMix)
rgmnMix < robustify(gmnMix, weight=0.1, mean=2)
rgmnMix
plot(rgmnMix)
nm < mixnorm(inf1=c(0.2, 0.5, 0.7), inf2=c(0.8, 2, 1), sigma=2)
plot(nm)
rnMix < robustify(nm, weight=0.1, mean=0, sigma=2)
rnMix
plot(rnMix)

Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.