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

From |
Timothy.Mak@iop.kcl.ac.uk |

To |
statalist@hsphsun2.harvard.edu |

Subject |
RE: st: GLLAMM question |

Date |
Mon, 22 Aug 2005 10:21:08 +0100 |

Hi, Are you doing this as a path model? Cos I can't imagine having 7 dichotomous level-2 variables that are not dummy variables. If it's a path model, can you re-express the problem in path model terms? Thanks. Tim "Hochman, Tsivia" <Tsivia.Hochman@med.va.gov> Sent by: owner-statalist@hsphsun2.harvard.edu 19/08/2005 17:26 Please respond to statalist@hsphsun2.harvard.edu To <statalist@hsphsun2.harvard.edu> cc Subject RE: st: GLLAMM question Thanks for the quick response. I am trying to run a multilevel logistic regression model. There are 8 dichotomous level-1 variables and 7 dichotomous level-2 variables. I would like to obtain odds ratios for all level-1 and level-2 variables. Presently, I am only able to obtain this information for the fixed effects. . eq lev_2: var1_2 var2_2 var3_2 var4_2 var5_2 var6_2 var7_2 . xi: gllamm outcome var1_1 var2_1 var3_1 var4_1 var5_1 var6_1 var7_1, i(idnum) geqs(lev2) -----Original Message----- From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Timothy.Mak@iop.kcl.ac.uk Sent: Friday, August 19, 2005 12:01 PM To: statalist@hsphsun2.harvard.edu Subject: Re: st: GLLAMM question Hi, the option eqs defines the factor loading for your random factors. The first loading is by default set to one. If you don't want the default, you specify the frload option. Exactly how to specify the model will depend on exactly what you want, so if you give more detail, I may be able to give more help. Tim "Hochman, Tsivia" <Tsivia.Hochman@med.va.gov> Sent by: owner-statalist@hsphsun2.harvard.edu 19/08/2005 16:37 Please respond to statalist@hsphsun2.harvard.edu To <statalist@hsphsun2.harvard.edu> cc Subject st: GLLAMM question I'm running a 2-level model using gllamm. For the level 1 variables (fixed effects), STATA produces coefficient estimates, standard errors and p-values. However, STATA only produces variance and covariance estimates for the level 2 variables (random effects). Is it possible to obtain coefficients estimates and p-values for the random effect component of this model? HLM6 generates such output. Thanks. * * 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/ * * 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/

**References**:**RE: st: GLLAMM question***From:*"Hochman, Tsivia" <Tsivia.Hochman@med.va.gov>

- Prev by Date:
**st: poverty decomposition** - Next by Date:
**st: Partitioning a Categorical Variable Based on Frequencies** - Previous by thread:
**RE: st: GLLAMM question** - Next by thread:
**st: RE: code to round a continuous variable to the nearest 1/10th** - Index(es):

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