When multiple serial intervals are provided, the posterior of R_t should be calculated using the approximation:
p(Rt|data) = 1/N sum(i=1 to N) p(R_t|data, w^(i))
where w^(i) are the samples of the serial interval. Posterior means and percentiles should be calculated based on this density and not on sampling from the gamma distribution.
When multiple serial intervals are provided, the posterior of R_t should be calculated using the approximation:
p(Rt|data) = 1/N sum(i=1 to N) p(R_t|data, w^(i))
where w^(i) are the samples of the serial interval. Posterior means and percentiles should be calculated based on this density and not on sampling from the gamma distribution.