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]

st: Mixing xtmixed and multiple imputation


From   "Gross, Alden L." <aldgross@jhsph.edu>
To   "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu>
Subject   st: Mixing xtmixed and multiple imputation
Date   Wed, 16 Oct 2013 12:23:53 -0400

Dear Statalist,

I am running a random intercept model using XTMIXED on multiply imputed data that were generated using the ice.ado command.

My question and concern: output in the Stata window is different from that which is saved in the returned r(table) matrix. Why does this happen? Which is the truth? Below is an example. The 0.807018 coefficient for intgrp_1 turns into .74022102 in r(table)!


mim: xtmixed pct_total intgrp_1 tiyear_1 intxtiy_1_1 || id: , cov(unstr)


Multiple-imputation estimates (xtmixed)                  Imputations =      11
Mixed-effects REML regression                            Minimum obs =    1404
                                                         Minimum dof =    64.9

------------------------------------------------------------------------------
   pct_total |     Coef.  Std. Err.     t    P>|t|    [95% Conf. Int.]     FMI
-------------+----------------------------------------------------------------
    intgrp_1 |   .807018   .780756    1.03   0.301   -.723245  2.33728   0.004
    tiyear_1 |   1.68917   .584153    2.89   0.005    .525116  2.85321   0.269
intxtiy_1_1 |   .312629   .801866    0.39   0.697   -1.27744   1.9027   0.220
       _cons |     49.64   .553024   89.76   0.000    48.5561  50.7239   0.005
-------------+----------------------------------------------------------------
   /lns1_1_1 |    2.0611   .040247                    1.98188  2.14033   0.128
    /lnsig_e |   1.90107   .031234                    1.83869  1.96345   0.289
------------------------------------------------------------------------------

mat list r(table)

r(table)[9,6]
          pct_total:   pct_total:   pct_total:   pct_total:    lns1_1_1:     lnsig_e:
           intgrp_1     tiyear_1  intxtiy_1_1        _cons        _cons        _cons
     b    .74022102    1.9004145    .19115065    49.642492    2.0656475    1.8845627
    se    .77570477    .49766392    .70280263    .54928689    .03721066    .02672613
     z    .95425612    3.8186705     .2719834    90.376255    55.512244    70.513858
pvalue    .33995401    .00013417    .78563478            0            0            0
   ll   -.78013238    .92501116   -1.1863172     48.56591    1.9927159    1.8321804
    ul    2.2605744    2.8758179    1.5686185    50.719075     2.138579     1.936945
    df            .            .            .            .            .            .
  crit     1.959964     1.959964     1.959964     1.959964     1.959964     1.959964
eform            0            0            0            0            0            0

Thank you in advance,

Alden Gross, PhD, MHS
Department of Epidemiology, JHSPH
Center on Aging and Health
Suite 2-700, 2024 E. Monument St.
Baltimore, MD  21231
aldgross@jhsph.edu



*
*   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/


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