prior_spike_and_slab creates a spike and slab prior distribution corresponding to the specification in Kuo and Mallick (1998) (see O'Hara and Sillanpää (2009) for further details). I.e., a prior distribution is multiplied by an independent indicator with values either zero or one.

prior_spike_and_slab(
  prior_parameter,
  prior_inclusion = prior(distribution = "spike", parameters = list(location = 0.5)),
  prior_weights = 1
)

Arguments

prior_parameter

a prior distribution for the parameter

prior_inclusion

a prior distribution for the inclusion probability. The inclusion probability must be bounded within 0 and 1 range. Defaults to prior("spike", parameters = list(location = 0.5)) which corresponds to 1/2 prior probability of including the slab prior distribution (but other prior distributions, like beta etc can be also specified).

prior_weights

prior odds associated with a given distribution. The value is passed into the model fitting function, which 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

return an object of class 'prior'.

See also

Examples

# create a spike and slab prior distribution
p1 <- prior_spike_and_slab(
   prior(distribution = "normal", parameters = list(mean = 0, sd = 1)),
   prior_inclusion = prior(distribution = "beta", parameters = list(alpha = 1, beta = 1))
)