One can put a Gaussian mixture prior on regression coefficients (George and McCulloch 1991), when data is not quite enough a cross like posterior can raise.
One can put a Gaussian mixture prior on regression coefficients (George and McCulloch 1991), when data is not quite enough a cross like posterior can raise.