Learning how to write a Zero Inflated Poisson model with stan?

[ignacio]

@andre.pfeuffer

However, y_sim needs some refinement. As I understand the zero-inflated hopefully correct,
with theta probability of a ‘0’ event a bernoulli is sampled.
If not “(1-theta) probability”, the possion is sampled.

Could you show me how to implement this? I based what I did on this post. This is how i’m reading my code:

1. Observation n will be zero with probability theta[n]: zero=bernoulli_rng(theta[n]);
2. If observation n is not zero, then it will draw a value from the Poisson: y_sim[n]=(1-zero)*poisson_log_rng(lambda_log[n])