negative AIC and BIC values in gls

classic Classic list List threaded Threaded
7 messages Options
Reply | Threaded
Open this post in threaded view
|

negative AIC and BIC values in gls

wudadan
Dear R users,

I obtained negative AIC and BIC and positive Loglik values in a gls model.
Is this normal? how should I interpret them? Thanks!

       AIC       BIC           logLik
  -659.978  -587.5541   345.989

Best
Gary

        [[alternative HTML version deleted]]

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Reply | Threaded
Open this post in threaded view
|

Re: negative AIC and BIC values in gls

Jeremy Miles-2
It's fine. Just interpret them as you would any other (lower is better).


On 22 August 2012 16:43, Gary Dong <[hidden email]> wrote:

> Dear R users,
>
> I obtained negative AIC and BIC and positive Loglik values in a gls model.
> Is this normal? how should I interpret them? Thanks!
>
>        AIC       BIC           logLik
>   -659.978  -587.5541   345.989
>
> Best
> Gary
>
>         [[alternative HTML version deleted]]
>
> ______________________________________________
> [hidden email] mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Reply | Threaded
Open this post in threaded view
|

Re: negative AIC and BIC values in gls

David Winsemius

On Aug 22, 2012, at 5:17 PM, Jeremy Miles wrote:

> It's fine. Just interpret them as you would any other (lower is  
> better).
>

And it is the printed logLik that is out of step here. log-likelihoods  
_should_ be negative.

--
David.

>
> On 22 August 2012 16:43, Gary Dong <[hidden email]> wrote:
>> Dear R users,
>>
>> I obtained negative AIC and BIC and positive Loglik values in a gls  
>> model.
>> Is this normal? how should I interpret them? Thanks!
>>
>>       AIC       BIC           logLik
>>  -659.978  -587.5541   345.989
>>
>> Best
>> Gary


David Winsemius, MD
Alameda, CA, USA

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Reply | Threaded
Open this post in threaded view
|

Re: negative AIC and BIC values in gls

Bert Gunter
In reply to this post by Jeremy Miles-2
Inline.



On Wed, Aug 22, 2012 at 5:17 PM, Jeremy Miles <[hidden email]> wrote:
> It's fine. Just interpret them as you would any other (lower is better).
>
I don't think so. I believe all 3 values are the negative of what they
should be. AIC is defined as -2*log(L) + k*{degrees of freedom for
model) . BIC is similar (different k's). This should be positive and,
as you said, the smaller the better. So if the signs are reversed,
models with lower absolute AIC and BIC -- bigger negative AIC and BIC
-- is what you want.

However, this really needs comment by an expert, which I ain't. I also
suggest you post to r-sig-mixed-models, since gls is an nlme function.

Cheers,
Bert

>
> On 22 August 2012 16:43, Gary Dong <[hidden email]> wrote:
>> Dear R users,
>>
>> I obtained negative AIC and BIC and positive Loglik values in a gls model.
>> Is this normal? how should I interpret them? Thanks!
>>
>>        AIC       BIC           logLik
>>   -659.978  -587.5541   345.989
>>
>> Best
>> Gary
>>
>>         [[alternative HTML version deleted]]
>>
>> ______________________________________________
>> [hidden email] mailing list
>> https://stat.ethz.ch/mailman/listinfo/r-help
>> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
>> and provide commented, minimal, self-contained, reproducible code.
>
> ______________________________________________
> [hidden email] mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.



--

Bert Gunter
Genentech Nonclinical Biostatistics

Internal Contact Info:
Phone: 467-7374
Website:
http://pharmadevelopment.roche.com/index/pdb/pdb-functional-groups/pdb-biostatistics/pdb-ncb-home.htm

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Reply | Threaded
Open this post in threaded view
|

Re: negative AIC and BIC values in gls

Ingmar Visser
In reply to this post by David Winsemius
> It's fine. Just interpret them as you would any other (lower is better).
>>
>>
> And it is the printed logLik that is out of step here. log-likelihoods
> _should_ be negative.
>

That is not quite the case; in models with small variances log-likelihoods
can easily become positive, consider eg:

> dnorm(0,0,0.1,log=T)
[1] 1.383647

hth, Ingmar



> --
> David.
>
>
>> On 22 August 2012 16:43, Gary Dong <[hidden email]> wrote:
>>
>>> Dear R users,
>>>
>>> I obtained negative AIC and BIC and positive Loglik values in a gls
>>> model.
>>> Is this normal? how should I interpret them? Thanks!
>>>
>>>       AIC       BIC           logLik
>>>  -659.978  -587.5541   345.989
>>>
>>> Best
>>> Gary
>>>
>>
>
> David Winsemius, MD
> Alameda, CA, USA
>
>
> ______________________________**________________
> [hidden email] mailing list
> https://stat.ethz.ch/mailman/**listinfo/r-help<https://stat.ethz.ch/mailman/listinfo/r-help>
> PLEASE do read the posting guide http://www.R-project.org/**
> posting-guide.html <http://www.R-project.org/posting-guide.html>
> and provide commented, minimal, self-contained, reproducible code.
>

        [[alternative HTML version deleted]]

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Reply | Threaded
Open this post in threaded view
|

Re: negative AIC and BIC values in gls

Bert Gunter
O course! And for the same reason, my stupid comment should be ignored.

-- Bert

On Thu, Aug 23, 2012 at 12:02 AM, Ingmar Visser <[hidden email]> wrote:

>> It's fine. Just interpret them as you would any other (lower is better).
>>>
>>>
>> And it is the printed logLik that is out of step here. log-likelihoods
>> _should_ be negative.
>>
>
> That is not quite the case; in models with small variances log-likelihoods
> can easily become positive, consider eg:
>
>> dnorm(0,0,0.1,log=T)
> [1] 1.383647
>
> hth, Ingmar
>
>
>
>> --
>> David.
>>
>>
>>> On 22 August 2012 16:43, Gary Dong <[hidden email]> wrote:
>>>
>>>> Dear R users,
>>>>
>>>> I obtained negative AIC and BIC and positive Loglik values in a gls
>>>> model.
>>>> Is this normal? how should I interpret them? Thanks!
>>>>
>>>>       AIC       BIC           logLik
>>>>  -659.978  -587.5541   345.989
>>>>
>>>> Best
>>>> Gary
>>>>
>>>
>>
>> David Winsemius, MD
>> Alameda, CA, USA
>>
>>
>> ______________________________**________________
>> [hidden email] mailing list
>> https://stat.ethz.ch/mailman/**listinfo/r-help<https://stat.ethz.ch/mailman/listinfo/r-help>
>> PLEASE do read the posting guide http://www.R-project.org/**
>> posting-guide.html <http://www.R-project.org/posting-guide.html>
>> and provide commented, minimal, self-contained, reproducible code.
>>
>
>         [[alternative HTML version deleted]]
>
> ______________________________________________
> [hidden email] mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.



--

Bert Gunter
Genentech Nonclinical Biostatistics

Internal Contact Info:
Phone: 467-7374
Website:
http://pharmadevelopment.roche.com/index/pdb/pdb-functional-groups/pdb-biostatistics/pdb-ncb-home.htm

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Reply | Threaded
Open this post in threaded view
|

Re: negative AIC and BIC values in gls

David Winsemius
In reply to this post by Ingmar Visser

On Aug 23, 2012, at 12:02 AM, Ingmar Visser wrote:

>
> It's fine. Just interpret them as you would any other (lower is  
> better).
>
>
> And it is the printed logLik that is out of step here. log-
> likelihoods _should_ be negative.
>
> That is not quite the case; in models with small variances log-
> likelihoods can easily become positive, consider eg:
>
> > dnorm(0,0,0.1,log=T)
> [1] 1.383647
>

We must be using different definitions of likelihood. (I do understand  
that densities can be greater than one when the domain is restricted.)

--
David.

> hth, Ingmar
>
>
> --
> David.
>
>
> On 22 August 2012 16:43, Gary Dong <[hidden email]> wrote:
> Dear R users,
>
> I obtained negative AIC and BIC and positive Loglik values in a gls  
> model.
> Is this normal? how should I interpret them? Thanks!
>
>       AIC       BIC           logLik
>  -659.978  -587.5541   345.989
>
> Best
> Gary
>
>
> David Winsemius, MD
> Alameda, CA, USA
>
>
> ______________________________________________
> [hidden email] mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>

David Winsemius, MD
Alameda, CA, USA


        [[alternative HTML version deleted]]

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.