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: mi and ice commands


From   [email protected] (Yulia Marchenko, StataCorp)
To   [email protected]
Subject   Re: st: mi and ice commands
Date   Thu, 10 Mar 2011 10:55:46 -0600

Aggie Chidlow <[email protected]> has a follow-up question about
obtaining the R-squared measure after -mi estimate: regress-:

> I am currently running "mi estimate: regress y x1 x2 x3" As I am used to
> reporting R2 for my results, I can't see it for the "mi estimate" command.
> Apologies for a silly question, but is R2 not reported in "mi estimate"

The -mi estimate- command is its own estimation command.  It reports
statistical results applicable to multiple-imputation (MI) analysis in
general.  Thus, the R-squared measure is not part of the -mi estimate- output.

Aggie can use the user-written command -mibeta- which is a wrapper for -mi
estimate: regress- that reports MI estimates of the standardized coefficients
and R-squared measures.  Aggie can type

  . findit mibeta

in Stata to locate and install this command.  For more information about how
to obtain multiple-imputation estimates of statistics other than coefficients
stored in -e(b)-, see the following FAQ:

  http://www.stata.com/support/faqs/stat/mi_combine.html


> In addition, would you mind telling me what are the main statistics that
> should be reported after running "mi estimate: regress y x1 x2 x3"?

The MI estimates of the R-squared and adjusted R-squared measures are often
reported with MI linear regression analysis (e.g., Harel 2009; Marshall et al.
2009).  Fractions of missing information associated with parameters of
interest should also be reported in any MI analysis.  These fractions (and
other information specific to MI estimates) can be obtained by specifying
the -vartable- option with -mi estimate-.


References:

Harel, O. 2009. The estimation of R2 and adjusted R2 in incomplete data sets
 using multiple imputation. Journal of Applied Statistics, 36: 1109-1118.

Marshall A., D. G. Altman, R. L. Holder, and P. Royston. 2009. Combining
 estimates of interest in prognostic modelling studies after multiple
 imputation: current practice and guidelines. BMC Medical Research Methodology
 9:57.


-- Yulia
[email protected]
*
*   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