Here is a grat new command to do this.
Yula Marchenko wrote a command, mibeta (findit mibeta), that does everything McDonald requested and more. It reports standardized beta weights and R-square when doing multiple imputation. It has an option to apply Fisher's z transformation in doing this given the upper limit issue of R-square and beta weights (under normal conditions). It can incorporate any of the mi estimate options such as reporting the relative efficiency for each parameter estimate for your number of imputed datasets, m, compared to an infinite number of imputed datasets. Here is a brief description Yulia has in the help file.
"By default, mibeta reports descriptive statistics (mean, median, minimum,
maximum, and the 25th and the 75th percentiles) of standardized coefficients,
R-squared and adjusted R-squared coefficients using the original scale. The
reported mean statistic obtained by averaging the estimates of these measures
over imputed data (that is, by applying Rubin's rules to the estimates in the
original metric) must be used with caution because the distribution of these
measures is not symmetric and may be far from Normal. Harel (2009) suggests to
use Fisher's z transformation for the R-squared measures to improve normality
(option fihserz). Marshall et al. (2009) recommend to look at quantiles (as
also reported by mibeta) of the distribution over the imputed datasets to
obtain more appropriate estimates."
On Wednesday, December 30, 2009, at 12:34PM, "McDonald, Catherine" <[email protected]> wrote:
>Hello-
>
>I am trying to figure out something with the mi commands in Stata 11. There is an option in the Linear Regression estimation to check off 'standardized beta coefficients' in reporting. But when I do this, nothing changes in my coefficients. I don't see any beta coefficients reported-the coefficients are the same as if I did not check it off. Can someone give me advice on this?
>
>Many Thanks-
>Catherine McDonald
>
>*
>* 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/
>
>
*
* 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/
