Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

RE: st: from Gauri (posting troubles) re positive log-likelihood


From   Gauri Khanna <gwkhanna@hotmail.com>
To   <statalist@hsphsun2.harvard.edu>
Subject   RE: st: from Gauri (posting troubles) re positive log-likelihood
Date   Sun, 7 Oct 2007 19:26:12 +0000

Dear Maarten,

Thank you for your reply. I have compared models and used the likelihood ratio test subsequently. 

I guess that the value of any single likelihood as it stands is not terribly important but is more useful when compared to another model. 

Thank you. 

Sincerely, 

Gauri 


> Date: Sun, 7 Oct 2007 18:43:46 +0100
> From: maartenbuis@yahoo.co.uk
> Subject: Re: st: from Gauri (posting troubles) re positive log-likelihood
> To: statalist@hsphsun2.harvard.edu
>
> --- Gauri wrote:
>> I get the following value for the Log likeihood = 97
>>
>> How can I interpret this?
>>
>> I understand that a positive log likelihood is not unknown when the
>> dependant variable is continuous which is true for my case.
>>
>> However, I am stuck with the interpretation and can't express it in
>> words.
>
> In case of a continuous depedent variable the log likelihood is not the
> log of a probability (which must remain below zero) but a probability
> density (from the probability density function). The probability
> density is not bounded to be less than or equal to one, so the log
> likelihood is not bounded to be negative. The value of one log
> likelihood is not so interpretable on its own, the real power is in
> comparing (nested) models (e.g. through a likelihood ratio test or
> BICs/AICs).
>
> Hope this helps,
> Maarten
>
> -----------------------------------------
> Maarten L. Buis
> Department of Social Research Methodology
> Vrije Universiteit Amsterdam
> Boelelaan 1081
> 1081 HV Amsterdam
> The Netherlands
>
> visiting address:
> Buitenveldertselaan 3 (Metropolitan), room Z434
>
> +31 20 5986715
>
> http://home.fsw.vu.nl/m.buis/
> -----------------------------------------
>
>
> ___________________________________________________________
> Want ideas for reducing your carbon footprint? Visit Yahoo! For Good http://uk.promotions.yahoo.com/forgood/environment.html
> *
> * For searches and help try:
> * http://www.stata.com/support/faqs/res/findit.html
> * http://www.stata.com/support/statalist/faq
> * http://www.ats.ucla.edu/stat/stata/

_________________________________________________________________
Connect to the next generation of MSN Messenger 
http://imagine-msn.com/messenger/launch80/default.aspx?locale=en-us&source=wlmailtagline
*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index