null.lim.cdf

stat

if true, - if stat is
  loglik_mean.brownianbridge, use the Anderson-Darling CDF
  #NO - if stat is var.brownianbridge, use the Cramer von
  Mises CDF - if stat is extremum.brownianbridge, use the
  Kolmogorov-Smirnov CDF #NO - if stat is
  ratio_loglik_extremum.brownianbridge, use the chi-square
  (3 freedom degrees) CDF - if stat is
  ratio_extremum.brownianbridge, use the Bay CDF else, use
  tabulated empirical CDF built on white noise process
  (length 10000)

forceUseECDF

Markov chain Monte Carlo burn-in based on "bridge" statistics, in the way of coda::heidel.diag, but including non asymptotic tabulated statistics.

null.lim.cdf: Asymptotic CDF for a given statistic

Description

Usage

Arguments