get_minimum_nw_IW.RdReturns minimum value for degrees of freedom such that the Inverted Wishart has a well-defined expected value.
get_minimum_nw_IW(p)
| p | number of variables |
|---|
minimum dof \(\nu_w\)
Uses (Press 2012) parametrization.
$$X \sim IW(\nu, S)$$
with \(S = pxp\) matrix, \(n_w > 2p\) (the degrees of freedom).
Then:
$$E[X] = S / (\nu_w - 2(p + 1))$$
Finally, the minimum dof \(\nu_w\) are: \(\nu_w = 2(p + 1) + 1\).
Press SJ (2012). Applied Multivariate Analysis: Using Bayesian and Frequentist Methods of Inference. Courier Corporation.
Other core functions:
bayessource-package,
make_priors_and_init(),
marginalLikelihood_internal(),
marginalLikelihood(),
mcmc_postproc(),
samesource_C(),
two.level.multivariate.calculate.UC()
Other Wishart functions:
diwishart_inverse_R(),
diwishart_inverse(),
diwishart(),
dwishart(),
riwish_Press(),
rwish()