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

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/

**References**:**st: from Gauri (posting troubles) re positive log-likelihood***From:*Kit Baum <baum@bc.edu>

**Re: st: from Gauri (posting troubles) re positive log-likelihood***From:*Maarten buis <maartenbuis@yahoo.co.uk>

- Prev by Date:
**Re: st: dropped observations in bootstrap** - Next by Date:
**Re: st: how do peoplelearn Stata?** - Previous by thread:
**Re: st: from Gauri (posting troubles) re positive log-likelihood** - Next by thread:
**st: re: checkreg3** - Index(es):

© Copyright 1996–2016 StataCorp LP | Terms of use | Privacy | Contact us | What's new | Site index |