convert_Multinom_Normal function

convert_Multinom_Normal

Calculate the parameters of the Dirichlet that best approximates the given log-Normal distribution. The approximation is the best in the sense that it minimizes the KL divergence from the log-Normal to the Dirichlet.

convert_Multinom_Normal(ft, Qt, parms = list())

Arguments

• ft: vector: A vector representing the means from the normal distribution.
• Qt: matrix: A matrix representing the covariance matrix of the normal distribution.
• parms: list: A list of extra known parameters of the distribution. Not used in this kernel.

Returns

The parameters of the conjugated distribution of the linear predictor.

See Also

Other auxiliary functions for a Multinomial outcome: convert_Normal_Multinom(), multnom_pred(), update_Multinom()