Truncation in manual


I just stumbled on page 55 over the explanation for truncation:

“The truncation at zero need not be specified as Stan only requires the density
up to a proportion”

To my understanding this statement is only correct for densities which are symmetric around zero (or whatever the truncation point is). The quote from above is right next to a truncated half-normal, so it is ok in the context, but if my understanding is correct we should maybe rephrase this, right?


Symmetry doesn’t matter. The point is that the missing
mass is constant.

What does matter is that the parameters of the density
be constants, not parameters of the model. Otherwise, the
amount of missing mass is no longer constant.

  • Bob