Dear Danay,
My feeling is you shouldn't exponentiate because the logs look more
normally distributed, but I'm not sure what the standard practice is.
Best,
Vladimir
2011/6/10 DANAY DIMA <[log in to unmask]>:
> Dear All,
>
> I would like to ask if when we perform further statistics on posterior means
> from DCM results (e.g. DCM.Ep.A) should we use the 'exp(DCM.Ep.A)' or just
> the 'DCM.Ep.A'?
>
> Thanks you in advance.
>
|