Thank you so much for your prompt help!! I wonder what does
poisson_log_lpmf(i | log_lambda) stand for (I did not see this function in the manual, only saw
Also, I think I did not clear it somehow, but what I am trying to find is the posterior distribution of
yT based on the product
p(x|y_T)p(y_T|y) (this final form is attained after some algebraic manipulations already). So shouldn’t I replace
i should be a column vector in order to make
A*i makes sense. But
i here is just a number, so I am not sure if you are discretizing everything basically??
Finally, could you elaborate on
To get the posterior predictive distribution of x just draw from a Poisson and use that realization multiplied by A to draw from a normal distribution with standard deviation sigma? I meant, didn’t you already do that with the
target variable? Also, in this case, what does my
model section in Stan become?