Hello,
I’ve been trying to fit some mixture models (Gaussian and Poisson). I am unsure of how to write the generated quantities block. I have seen some use cases in the discourses as well as tried to adapt the code in the Stan manual but have not had meaningful results.
I put the code for my Gaussian mixture model below, I would greatly appreciate some help with writing the “generated quantities” block.
mix_model <- "
data {
int<lower = 0> N;
vector[N] y;
}
parameters {
ordered[2] mu;
real<lower=0> sigma[2];
real<lower=0, upper=1> theta;
}
model {
sigma ~ normal(0, 2);
mu[1] ~ normal(70, 10);
mu[2] ~ normal(110,10);
theta ~ beta(5, 5);
for (n in 1:N){
target += log_mix(theta,
normal_lpdf(y[n] | mu[1], sigma[1]),
normal_lpdf(y[n] | mu[2], sigma[2]));
}
}
"
Please let me know if I can provide more information/data etc to make my question clearer.
Additionally a question about Stan manual (version 2.17.0),
on page 194, for a mixture model with K mixture components the following code is given for the parameters block;
parameters {
simplex[K] theta; // mixing proportions
ordered mu[K]; // locations of mixture components
vector<lower=0>[K] sigma; // scales of mixture components
}
I think the [K] should be right after “ordered”, because Stan threw up an error, unless I am wrong, some clarification would help.