Open cschwarz-stat-sfu-ca opened 5 years ago
We use the dgamma prior for variance components. However, Speigelhalter and others argue that this is holdover from the pre-computer era and a better prior may be uniform(?).
This should be fixed in all of the modules.
We use the dgamma prior for variance components. However, Speigelhalter and others argue that this is holdover from the pre-computer era and a better prior may be uniform(?).
This should be fixed in all of the modules.