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

From |
Shuaizhang Feng <fengsz@yahoo.com> |

To |
statalist@hsphsun2.harvard.edu |

Subject |
Re: st: RE: random effects, fixed effects and mixed effects models |

Date |
Thu, 13 Oct 2005 20:24:48 -0700 (PDT) |

Thanks. I guess what I am really looking for is an approach that could use both within and between variations to estimate the model but without assuming that the "random effect" is orthogonal to explanatory variables. Is this possible with STATA? Thanks a lot. --- Steve Stillman <stillman@motu.org.nz> wrote: > GLLAMM and XTMIXED are capable of estimating a > wide-range of models, but in general, these models > are in the random effects family and assume that c > is orthogonal to x. I am not sure what you mean by > "I have too few observations". There is nothing in > particular about fixed effects models that require > more observations than OLS or random effects > (assuming that you have at least two time periods of > data on many/most individuals/firms). > > Steve > > -----Original Message----- > From: owner-statalist@hsphsun2.harvard.edu > [mailto:owner-statalist@hsphsun2.harvard.edu]On > Behalf Of Shuaizhang > Feng > Sent: Friday, October 14, 2005 8:17 AM > To: statalist@hsphsun2.harvard.edu > Subject: st: random effects, fixed effects and mixed > effects models > > > Dear all: > > I have a general question on what to choose with a > panel data model like this: > > y(it)=x(it)b+c(i)+v(it) > > If I use xtreg to estimate this, when using random > effects models (re), then I am assuming that c is > orthogonal to x, right? But that is some assumption > I > am reluctant to make. I do not want to use fixed > effects estimates too because I have too few > observations. > > So could I use GLLAMM or XTMIXED. I suppose in those > c > is not assumed to be orthogonal to x. Is that true? > also, if both GLLAMM and XTMIXED work, which one > should I choose? I may have to use a two level mixed > effects model like this one. > > y(ijt)=x(ijt)b+c1(ij)+c2(i)+v(ijt). > > Thanks a lot. > > SZ > > > > > __________________________________ > Yahoo! Mail - PC Magazine Editors' Choice 2005 > http://mail.yahoo.com > * > * 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/ > > > * > * 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/ > __________________________________ Yahoo! Mail - PC Magazine Editors' Choice 2005 http://mail.yahoo.com * * 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/

**Follow-Ups**:**Re: st: RE: random effects, fixed effects and mixed effects models***From:*Sean <asmileguo@gmail.com>

**References**:**st: RE: random effects, fixed effects and mixed effects models***From:*"Steve Stillman" <stillman@motu.org.nz>

- Prev by Date:
**st: RE: Re: Marginal effect after xtlogit, fe** - Next by Date:
**st: bootstrap ROC areas (optimism) on logistic model** - Previous by thread:
**st: RE: random effects, fixed effects and mixed effects models** - Next by thread:
**Re: st: RE: random effects, fixed effects and mixed effects models** - Index(es):

© Copyright 1996–2015 StataCorp LP | Terms of use | Privacy | Contact us | What's new | Site index |