Calculate posterior parameter for the Normal-Gamma, assuming that the observed values came from a Normal model from which the prior distribution for the mean and the precision have joint distribution Normal-Gamma
update_NG2(conj.param, ft, Qt, y, parms = list())
Arguments
conj.param: list: A vector containing the parameters of the Normal-Gamma (mu0,c0,alpha,beta).
ft: numeric: A vector representing the means from the normal distribution. Not used in the default method.
Qt: matrix: A matrix representing the covariance matrix of the normal distribution. Not used in the default method.
y: numeric: A vector containing the observations.
parms: list: A list of extra known parameters of the distribution. Not used in this kernel.