Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


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

Re: st: Gllapred and Level 2 Residuals


From   Stas Kolenikov <[email protected]>
To   [email protected]
Subject   Re: st: Gllapred and Level 2 Residuals
Date   Fri, 14 May 2010 22:01:49 -0500

You have numeric issues and/or identification problems in your model:
the random intercepts and slopes are pretty much perfectly correlated
(although negatively). That may be enough to break it down.

With normal data, you might be better off with -xtmixed-. You will
certainly get your estimates faster.

On Fri, May 14, 2010 at 6:15 PM, Jielu Lin <[email protected]> wrote:
> Dear all,
>
> I estimated a random coefficient model using gllamm and obtained level
> 2 residuals using gllapred with ustd option. However, there were no
> values for the residuals of the random slope (rz_iadl_q~m2 in the
> output below) . Here is my output:
>
> . gllamm iadl age agesq , i(id) eqs(inter slope) nip(12) nrf(2) adapt
> from(a) copy
> number of level 1 units = 580
> number of level 2 units = 145
> Condition Number = 390550.68
> log likelihood = -1081.5765
> ------------------------------------------------------------------------------
>      totiadl |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
> -------------+----------------------------------------------------------------
>         age |  -1.663252   .2012228    -8.27   0.000    -2.057641   -1.268862
>       agesq |   .0118635    .001325     8.95   0.000     .0092667    .0144604
>        _cons |   58.63494   7.653772     7.66   0.000     43.63382    73.63606
> ------------------------------------------------------------------------------
> Variance at level 1
> ------------------------------------------------------------------------------
>   1.3431285 (.10917208)
> Variances and covariances of random effects
> ------------------------------------------------------------------------------
> ***level 2 (id)
>     var(1): 125.06127 (29.45089)
>     cov(2,1): -1.8267617 (.4144318) cor(2,1): -.99814552
>     var(2): .02678263 (.00585232)
> ------------------------------------------------------------------------------
>
> . gllapred rz_iadl_q_r, ustd
> (means and standard deviations will be stored in rz_iadl_q_rm1
> rz_iadl_q_rs1 rz_iadl_q_rm2 rz_iadl_q_rs2 )
> Non-adaptive log-likelihood: -1081.3564 -1081.6149 -1081.5765
> -1081.5765   log-likelihood:-1081.5765
>
> . su  rz_iadl_q_rm1 rz_iadl_q_rs1 rz_iadl_q_rm2 rz_iadl_q_rs2
>     Variable |       Obs        Mean    Std. Dev.       Min        Max
> -------------+--------------------------------------------------------
> rz_iadl_q~m1 |       580   -.0277825    1.043866  -3.951646   2.430058
> rz_iadl_q~s1 |       580    9.382843    .9076038   8.122407   10.82412
> rz_iadl_q~m2 |         0
> rz_iadl_q~s2 |         0
>
> Thanks everyone in advance!
> Jielu Lin
>
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/
>



-- 
Stas Kolenikov, also found at http://stas.kolenikov.name
Small print: I use this email account for mailing lists only.

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index