| ![](/includes/contimages/spacer.gif) |
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
st: mim: reg command
I notice that when you use the mim command (available from SSC) you
don't get an R^2 statistic, e.g.
. mim: reg total black white worker exceed ltplan
Multiple-imputation estimates (regress) Imputations = 5
Linear regression Minimum obs = 4519
Minimum dof = 1000.0
------------------------------------------------------------------------------
total | Coef. Std. Err. t P>|t| [95% Conf. Int.] MI.df
-------------+----------------------------------------------------------------
black | -2342.99 29996.3 -0.08 0.938 -61205.9 56519.9 1000.0
white | 36197.6 22577.7 1.60 0.109 -8107.44 80502.6 1000.0
worker | 31077.2 17928.7 1.73 0.083 -4105.04 66259.4 1000.0
exceed | 34791.1 19359.1 1.80 0.073 -3198.07 72780.3 1000.0
ltplan | 36141.7 16114.5 2.24 0.025 4519.51 67763.8 1000.0
_cons | -43537.8 27718.2 -1.57 0.117 -97930.3 10854.7 1000.0
------------------------------------------------------------------------------
Would it be legit to average the R^2 from the separate imputations,
or to otherwise compute it? If not, is there a handy reference
somewhere explaining why R^2 is not a legit statistic when doing
multiple imputations? I just know that some reviewer is going to ask
where the R^2 statistic is!
-------------------------------------------
Richard Williams, Notre Dame Dept of Sociology
OFFICE: (574)631-6668, (574)631-6463
HOME: (574)289-5227
EMAIL: [email protected]
WWW: http://www.nd.edu/~rwilliam
*
* 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/