Dear Stan Board,

I am trying to fit a spatially varying parameter model and am running into some trouble with Kronecker products. This is the general form of the distribution I am trying to specify.

(\theta_{1s}, \theta_{2s}, \theta_{3s}) ~ MVN(**1** \otimes \mu_s, H(\phi) \otimes \Sigma).

I am trying to figure out the best way to code the mean vector. It looks like you can’t use the rep function with a vector (e.g., `rep(mu_s, 10)`

). I also looked at using `append_row`

but it is not obvious how to use this iteratively using replicates of the same vector. Does anyone know of a simple way to replicate \mu_s for varying length of **1**?

Colin