siarelicit: Elicit prior parameters for the Dirichlet distribution.
Description
Allows users to enter mean estimated proportions and a standard deviation term so that useful prior distributions
can be enterred into the siarmcmcdirichletv4 function.
Usage
siarelicit(siardata)
Arguments
siardata
A list containing some or all of the following parts: targets, sources, corrections,
PATH, TITLE, numgroups, numdata, numsources, numiso, SHOULDRUN, GRAPHSONLY, EXIT, and output. For more details
of these inputs see the siarmenu function. Details
Uses standard results from the Dirichlet distibution to turn the estimated mean proportions $M[i]$ and a
variance term $V[1]$ to give the estimated parameters $a[i]$ via:
$$a_i = M_i \left( \frac{M_1 (1-M_1)}{V_1} - 1 \right)$$
Note that $V$ can be given for any of the $k$ sources.