Hi, I’m currently playing around with mixture models, using a Dirichlet distribution for the weights of the mixture components. Is there a way to put priors on those weights or to limit the range for the individual weights to some parameter space? Any help is much appreciated.

Fixed or constrained weights in mixture models

Aniket_Pancharia March 8, 2023, 7:49pm 2

Hello @Ultraslan82 !
as far as my knowledge goes we cant put an upper limit to priors in a Dirichlet distribution however if you find a solution to that do let me know too

Topic		Replies	Views
Mixture Model Dirichlet Questions	7	3054	June 1, 2018
Using a mixture of multivariate normal distributions as a prior Questions	16	1826	March 12, 2019
4 lines basic mixture model outputs wrong results Questions	0	394	March 12, 2020
How to bound a Dirichlet? Questions	4	640	March 15, 2018
Which Distribution for Constrained Parameter vector Sum Questions	2	432	August 10, 2018

Fixed or constrained weights in mixture models

Related topics