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: Mean and standard deviation of multiply-imputed values?
From
Stas Kolenikov <[email protected]>
To
[email protected]
Subject
Re: st: Mean and standard deviation of multiply-imputed values?
Date
Mon, 27 Sep 2010 12:41:51 -0500
On Sat, Sep 25, 2010 at 4:39 PM, Art Burke
<[email protected]> wrote:
> If I have a multiply-imputed variable x1, I can get its mean and standard error from ....
>
> mi estimate: mean x1
>
> Can I get the standard deviation also (without any hand calculation)?
No, that's not what multiple imputation is expected to do; it only
gives point estimates, i.e., means, proportions, regression
coefficients. Nonlinear operations like sums of squares are outside of
what's sensible with -mi-. Hence what you can do is to create the
square of your x, register it, estimate its mean using -mi-, and then
obtain the point estimate of the variance of x as E[x^2] - ( E[x] )^2.
Still hand calculations though.
--
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/