complexityPrior {beast} | R Documentation |
Complexity prior distribution
Description
This function computes the complexity prior distribution on the number of change-points, defined as f(\ell) = P(\ell_n = \ell)\propto e^{-\alpha\ell\log(bT/\ell)}, a, b > 0; \ell = 0,1,2,\ldots
. Note that this distribution has exponential decrease (Castillo and van der Vaart, 2012) when b>1+e
, so we set b=3.72
.
Usage
complexityPrior(Lmax = 20, gammaParameter, nTime)
Arguments
Lmax |
maximum number of change-points (default = 20). |
gammaParameter |
positive real number, corresponding to |
nTime |
positive integer denoting the total number of time-points. |
Value
logPrior |
Prior distribution values in the log-scale. |
Author(s)
Panagiotis Papastamoulis
References
Castillo I. and van der Vaart A (2012). Needles and Straw in a Haystack: Posterior concentration for possibly sparse sequences. The Annals of Statistics, 40(4), 2069–2101.
[Package beast version 1.1 Index]