**announced** that Statalist is moving from an email list to a **forum**. The old list will shut down on April 23, and its replacement, **statalist.org** is already up and running.

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

From |
Stuart Buck <stuartbuck@gmail.com> |

To |
statalist@hsphsun2.harvard.edu |

Subject |
Re: st: confusion |

Date |
Sun, 5 May 2013 19:24:33 -0500 |

The Hausman test does not tell you to use fixed vs. random effects. Instead, if a Hausman test shows significant differences between the models, it's simply telling you that there are between effects that need to be accounted for via a mean group variable. See Bafumi/Gelman, http://www.stat.columbia.edu/~gelman/research/unpublished/Bafumi_Gelman_Midwest06.pdf, and especially Bell/Jones, http://polmeth.wustl.edu/media/Paper/FixedversusRandom_1_2.pdf . If you include the relevant group means, the fixed effects and random effects estimates for all the other variables will be identical. On Sat, May 4, 2013 at 10:25 AM, montaha <muntahakhader@hotmail.com> wrote: > hello everybody, > i need your help guys > i had to test for fixed and random effects using STATA. > I got the following tables which shows that > In "fixed effects" output the signs of the coefficient are totally different > from anticipated signs in theory, additionally some of them are > insignificant. > But in Random effect output the signs are true and they are significant. > when i did Hausman test.. the prob>chi2 was 0.000 which indicates that i > shall use Fixed effect > but as i have mentioned the signs are not as expected and they are not > significant... > the output > > <http://statalist.1588530.n2.nabble.com/file/n7580362/1_1.png> > <http://statalist.1588530.n2.nabble.com/file/n7580362/2_2.png> > <http://statalist.1588530.n2.nabble.com/file/n7580362/3_3.png> > > thank u i advance > > > > -- > View this message in context: http://statalist.1588530.n2.nabble.com/confusion-tp7580362.html > Sent from the Statalist mailing list archive at Nabble.com. > * > * For searches and help try: > * http://www.stata.com/help.cgi?search > * http://www.stata.com/support/faqs/resources/statalist-faq/ > * http://www.ats.ucla.edu/stat/stata/ -- Stuart Buck http://about.me/stuartbuck 479-200-2750 (cell) * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/faqs/resources/statalist-faq/ * http://www.ats.ucla.edu/stat/stata/

**References**:**st: confusion***From:*montaha <muntahakhader@hotmail.com>

- Prev by Date:
**st: Instrumental Variables approach using -xt- for three-level growth model** - Next by Date:
**Re: st: Time-series operators and -outreg-** - Previous by thread:
**st: confusion** - Next by thread:
**st: Reference for clustered standard errors in Stata** - Index(es):