Thanks again for the very detailed explanation.
mu = runner_mean[runner_idx] + year_mean[year_idx]
I didn’t know how to pass the mu to the observations. If i just pass mu or mu[index]. Is it something like the following?
Y_obs = pm.Normal(“Runner”, mu=mu,sigma=sig,observed=data.times)
And if i want to plot a specific mean for the first combination of runner+year)
use something like:
m = pm.Deterministic(‘m’,mu)
az.plot_trace(trace.posterior.m.sel(m_dim_0=0))
for the second
az.plot_trace(trace.posterior.m.sel(m_dim_0=1))