Skip to contents

Penalized complexity (PC) prior

Source: R/priors.R
prior_pc.Rd

Specify a PC prior for a positive parameter (typically a standard deviation). PC priors shrink toward simpler models by penalizing deviation from a base model.

Usage

prior_pc(U = 1, alpha = 0.01)

Arguments

U

Upper bound. P(x > U) = alpha.

alpha

Tail probability. Default 0.01.

Value

A ratiod_prior object

Details

The PC prior is specified via: P(sigma > U) = alpha

  • U is the upper bound you consider "large"

  • alpha is the probability of exceeding it

This implies an exponential prior with rate = -log(alpha) / U.

References

Simpson, D., Rue, H., Riebler, A., Martins, T. G., & Sorbye, S. H. (2017). Penalising model component complexity: A principled, practical approach to constructing priors. Statistical Science, 32(1), 1-28.

Examples

prior_pc(U = 1, alpha = 0.01)    # P(sigma > 1) = 0.01
#> PC prior: P(x > 1.00) = 0.010
#>   => Exponential(4.605)
prior_pc(U = 0.5, alpha = 0.05)  # Tighter, P(sigma > 0.5) = 0.05
#> PC prior: P(x > 0.50) = 0.050
#>   => Exponential(5.991)