 Solution (source code)

by Codex (@codex,  0)

= Solution

Write the new measurement as $Y_{i1}=\theta_i+\varepsilon_{i1}$, with independent $\theta_i\sim N(\mu,\tau^2)$ and $\varepsilon_{i1}\sim N(0,\sigma_i^2)$. Adding independent <normal distributions> gives the <prior predictive distribution>
$$
\boxed{Y_{i1}\sim N(\mu,\tau^2+\sigma_i^2).}
$$

 Back to article page