I’ve made some fake data and I’m trying to recover the parameters with a Stan model as a training exercise. I’m not getting what I want at all and I’d really appreciate if someone could help me out. The fake dataset as shown in the code below is just two Dirichlet mixtures with \alpha's (1,2,10) an…

[image] emiruz: for some reason theta1 and theta2 are not estimated at all. I don’t see this issue. When I run your code I get mean se_mean sd 2.5% 25% 50% 75% 97.5% n_eff Rhat theta1[1] 0.28 0.00 0.08 0.12 0.25 0.27 0.29 0.51 1098 1.00 theta1[2] 0.32 0.00 0…

Recovering parameters from fake Dirichlet data

Modeling

andrjohns July 30, 2019, 2:13pm 6

For the label-switching, a possible approach could be an ordered simplex: Ordered Simplex

But you should also read Michael Betancourt’s case study on label-switching for more info

1 Like

Topic		Replies	Views
Recovering parameters of fake model Modeling	1	655	April 27, 2022
Problem recovering parameters from mixture Rasch with Dirichlet prior Modeling	1	379	April 14, 2020
Having trouble recovering some model parameters when fitting the simulated data Modeling techniques , fitting-issues , irt	4	1069	June 11, 2022
Having trouble recovering known parameters from simulated data Modeling cmdstanpy	2	636	June 4, 2021
Unable to recover simulated parameters Modeling specification	10	1831	February 11, 2018

Recovering parameters from fake Dirichlet data

Related topics