prior_weightfunction {BayesTools}R Documentation

Creates a prior distribution for a weight function

Description

prior_weightfunction creates a prior distribution for fitting a RoBMA selection model. The prior can be visualized by the plot function.

Usage

prior_weightfunction(distribution, parameters, prior_weights = 1)

Arguments

distribution

name of the prior distribution. The possible options are

"two.sided"

for a two-sided weight function characterized by a vector steps and vector alpha parameters. The alpha parameter determines an alpha parameter of Dirichlet distribution which cumulative sum is used for the weights omega.

"one.sided"

for a one-sided weight function characterized by either a vector steps and vector alpha parameter, leading to a monotonic one-sided function, or by a vector steps, vector alpha1, and vector alpha2 parameters leading non-monotonic one-sided weight function. The alpha / alpha1 and alpha2 parameters determine an alpha parameter of Dirichlet distribution which cumulative sum is used for the weights omega.

parameters

list of appropriate parameters for a given distribution.

prior_weights

prior odds associated with a given distribution. The model fitting function usually creates models corresponding to all combinations of prior distributions for each of the model parameters, and sets the model priors odds to the product of its prior distributions.

Value

prior_weightfunction returns an object of class 'prior'.

See Also

plot.prior()

Examples

p1 <- prior_weightfunction("one-sided", parameters = list(steps = c(.05, .10), alpha = c(1, 1, 1)))

# the prior distribution can be visualized using the plot function
# (see ?plot.prior for all options)
plot(p1)


[Package BayesTools version 0.2.17 Index]