We are given the model $$ \begin{align*} y_{ij} & \sim \mathsf{Normal}(\alpha_j + \beta x_i, \sigma^2)\\ \alpha_j & \sim \mathsf{Normal}(\gamma_0 + \gamma_1 u_j, \tau^2) \end{align*} $$ with priors $$ \begin{align*} \beta, \gamma_0, \gamma_1 & \sim \mathsf{Normal}(0, 0.5^2)\\ \sigma, \tau & \sim \mathsf{Normal}_+(0, 0.25^2) \end{align*} $$ where we have $j$ counties and $i$ observations. I'm having trouble specifying all of my priors for rstanarm. What I have so far is


fit_lmer <- stan_glmer(log_radon ~ floor + (1 | county), 
                       data = radon,
                       prior_intercept = normal(0, 0.5),
                       prior_aux = normal(0, 0.25)) 

I'm just not entirely sure I properly specified everything.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.