Hi,
I am doing Bayesian Calibration for a model which has 24 parameters using MCMC.
I get a sample of 100,000 points from the posterior pdf. Let's say, It is a
dataframe named as "BC_output", with 24 parameters (V1,V2,V24)
I would like to show the Maximum a Posteriori (MAP)parameter value
Further discussed on r-sig-mixed-models
Rainer
On 22/08/12 17:04, Bert Gunter wrote:
Oops -- missed that. OTOH, my reply demonstrates the value of the
mixed models list recommendation.
-- Bert
On Wed, Aug 22, 2012 at 7:55 AM, Rainer M Krug wrote:
On 22/08/12 16:36, Bert Gunter wrote:
Mode
Oops -- missed that. OTOH, my reply demonstrates the value of the
mixed models list recommendation.
-- Bert
On Wed, Aug 22, 2012 at 7:55 AM, Rainer M Krug wrote:
> On 22/08/12 16:36, Bert Gunter wrote:
>>
>> Models with different fixed effects estimated by REML cannot be
>> compared by anova.
>
On 22/08/12 16:36, Bert Gunter wrote:
Models with different fixed effects estimated by REML cannot be
compared by anova.
I have seen that much in "Modern Applied Statistics in S", and therefore have chosen the
model = "ML"
In future, please post questions on mixed effects models on the
r-si
Models with different fixed effects estimated by REML cannot be
compared by anova.
In future, please post questions on mixed effects models on the
r-sig-mixed-effects mailing lists. You're likely to receive more
informative replies there, too.
-- Bert
On Wed, Aug 22, 2012 at 7:23 AM, Rainer M Kr
5 matches
Mail list logo
