Changed conjugate prior of Normal/LogNormal distributions to be the
NormalInverseGamma distribution from a combination of the
Inverse Gamma distributions. This distribution is bivariate and gives us a 2d estimate for both
sig_sq. The params for this distribution are
beta and are different from the old priors that Normal/LogNormal were expecting.
Various doc changes to illustrate these changes and new expectations
rename to retrieve and rename posteriors from your
Mostly useful in conjunction with
combine in order to quickly chain together several
Standardized prior parameters to have the same arguments as the
bayesTest(distribution = c('normal', 'lognormal'))
This is a breaking change
bayesTest$inputs$distributionto be consistent
Band not include the parameter name
B_datain inputs are now always lists by default to make
combinework more simply
bayesTestworks internally. Dispatch per distribution is now only related to how the posterior is calculated.
deployBanditto turn your
bayesTestobject into a Bayesian multi*armed bandit and deploy as a JSON API respectively.
Added programmatic capabilities on top of existing interactive uses for
plot generic function
You can now assign
plot(bayesTestObj) to a variable and not have it automatically plot.
Added Posterior Expected Loss to output of
Supports the risk of choosing ‘B’ over ‘A’ (ordering is important) and makes more sense if A > B currently in the test
plot generics are now explicitly
ggplot objects and can be modified as such
You can input your own titles/axis labels/etc if the defaults don’t fit your use case
combinetests as needed