suggest.burn {BoomSpikeSlab} | R Documentation |
Suggest Burn-in
Description
Suggest a burn-in period for a Bayesian neural network model.
Usage
SuggestBurn(model)
Arguments
model |
An object inheriting from class |
Details
See SuggestBurnLogLikelihood
for details of the on how
the burn-in period is suggested. In this case the negative the
residual standard deviation is used as a proxy for log likelihood.
Value
A non-negative integer less than the number of MCMC draws.
Author(s)
Steven L. Scott
See Also
[Package BoomSpikeSlab version 1.2.6 Index]