Это изучается в композиционном анализе данных, есть книга Aitchison: Статистический анализ композиционных данных .
Sn={(x1,…,xn+1)∈Rn+1:x1>0,…,xn+1>0,∑i=1n+1xi=1}.
nxx~x=(x1,…,xn+1)↦(log(x1/x~),…,log(xn/x~)Rn, so have an inverse which I leave to you to calculate (There are also other versions of this transformation that can be used, which has maybe better mathematical properties, more about that later).
Now you can take a normal (or whatever) distribution defined on Rn and use this inverse transformation to define a distribution on the simplex. The possibilities are limitless, for each and every multivariate distribution on Rn we get a distribution on the simplex.
I will augment this post later with some examples, and more details on log-ratio transforms.