Fitting a model to predictions from another unavailable model

If the poisson assumption is good then the means give you the variances. You could sum over all plausible unobserved poisson values and add that to the log density rather than using the rng.