The order of estimation

Modeling
1

In stan, if the order of estimation is specified as in the image, is there any way to achieve that?

image

For simplicity, we will use the following example
1, σ|alpha,beta
2, alpha|σ,bata
3, bata|alpha,σ
I think the algorithm is as follows, but is such a representation possible in stan?

model {
  vector[N] mu = alpha + beta * x;
  y ~ normal(mu, sigma); 
}
2

This issue of order-of-estimation is specific to Gibbs sampling, which is not the same thing as the HMC sampling that Stan uses.

3 Likes
3

OMG, this elementary question was caused by my lack of understanding.
Thank you very much for your kindness.