Predictive PDF — predictive

predictive_pdf() evaluates the predictive PDF of the model specified by a DataDistribution dist and Prior at the given stage-one outcome.

predictive_pdf(dist, prior, x1, n1, ...)

# S4 method for DataDistribution,PointMassPrior,numeric
predictive_pdf(dist, prior, x1, n1, ...)

# S4 method for DataDistribution,ContinuousPrior,numeric
predictive_pdf(
  dist,
  prior,
  x1,
  n1,
  k = 10 * (prior@support[2] - prior@support[1]) + 1,
  ...
)

Arguments

dist: a univariate distribution object
prior: a Prior object
x1: stage-one test statistic
n1: stage-one sample size
...: further optional arguments
k: number of pivots for crude integral approximation

Value

numeric, value of the predictive PDF

Examples

predictive_pdf(Normal(), PointMassPrior(.3, 1), 1.5, 20) # ~.343
#> [1] 0.3426953

tmp <- ContinuousPrior(function(x) dunif(x, .2, .4), c(.2, .4))
predictive_pdf(Normal(), tmp, 2, 20)
#> [1] 0.2302199