I'm sorry, I must be getting confused. My message from yesterday (question 1) asked if the objective function value in the .par file was the post-integration marginal likelihood, to which you responded in the negative, but now it seems as if you're saying what I had originally postulated is true. Maybe it's just a terminology difference. It is my understanding that the .par file contains the optimum loglikelihood value obtained, where the loglikelihood value is marginalized over the random effects. Perhaps there is some miscommunication? I think we're talking about the same thing.<div>
<div><br></div><div>Since the marginal likelihood is available in the .par file, is there a way to output the conditional likelihood following optimization?</div><div><br></div><div><br></div><div><br></div><div>Thanks again,</div>
<div><br></div><div>Chris</div><div><br></div><div><br></div><div><br>-----------------------------<br>Chris Gast<br><a href="mailto:cmgast@gmail.com">cmgast@gmail.com</a><br>
<br><br><div class="gmail_quote">On Tue, Feb 22, 2011 at 8:01 AM, H. Skaug <span dir="ltr"><<a href="mailto:hskaug@gmail.com">hskaug@gmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">
The marginal likelihood (= integrated likelhood = likelihood) at the<br>
fit is given in the par file as<br>
"Objective function value =".<br>
<br>
Hans<br>
<div><div></div><div class="h5"><br>
<br>
On Tue, Feb 22, 2011 at 4:36 PM, Chris Gast <<a href="mailto:cmgast@gmail.com">cmgast@gmail.com</a>> wrote:<br>
> Thank you, Hans.<br>
> Do you know of a way to output the optimum marginal likelihood from ADMB, or<br>
> is this available only internally?<br>
><br>
> -----------------------------<br>
> Chris Gast<br>
> <a href="mailto:cmgast@gmail.com">cmgast@gmail.com</a><br>
><br>
><br>
> On Tue, Feb 22, 2011 at 1:13 AM, H. Skaug <<a href="mailto:hskaug@gmail.com">hskaug@gmail.com</a>> wrote:<br>
>><br>
>> Hi,<br>
>><br>
>> ><br>
>> > 1) Is it correct to say that the objective function value returned in<br>
>> > the<br>
>> > .par value of a RE model is the approximation to the marginal<br>
>> > likelihood,<br>
>> > post-integration?<br>
>><br>
>> No<br>
>><br>
>> > 2a) Is it correct to obtain the conditional likelihood value by<br>
>> > re-evaluating the likelihood function at the MLEs (for fixed parameters)<br>
>> > and<br>
>> > empirical Bayes estimates of REs?<br>
>><br>
>> Yes, if I read you right.<br>
>><br>
>> > 2b) If 2a is true, can this value be obtained in a single ADMB<br>
>> > model-fit, or<br>
>> > do I need to first fit the model, then re-run the model using the<br>
>> > optimum<br>
>> > values from the previous run and output an initial likelihood value to<br>
>> > get<br>
>> > the conditional likelihood?<br>
>><br>
>> Single run. The .par file contains MLEs of fixed effects and empirical<br>
>> bayes<br>
>> estimtes of random effects. The MLEs are approximates, but that does<br>
>> not concern you conceptual disucssion.<br>
>><br>
>> Hans<br>
><br>
><br>
</div></div></blockquote></div><br></div></div>