Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: Multiple imputation: negative imputed values.

From   Stas Kolenikov <>
Subject   Re: st: Multiple imputation: negative imputed values.
Date   Thu, 30 Dec 2010 09:43:13 -0600

On Thu, Dec 30, 2010 at 9:29 AM, Khoi Dinh To <> wrote:
> 1. A few original variables do not have negative values (for example, faculty salaries, the number of bachelor's degrees produced...).  However, after I did 5 imputations (5 iterations), I found out that there were negative imputed values in those variables, too.  Are those negative values acceptable (because the original data went through 5 iterations)?  If they are not, then is there any measure to correct them?  I am asking these questions because I think I will have to present the descriptives of the imputed data, and I want to make sure the negative imputed values are acceptable to the public.

Read the FAQ on posting efficient statalist questions. At the very
least, you might want to provide your syntax; better yet, provide an
example with auto.dta that would create a situation similar to yours.
I suspect you are not using the right imputation model.

> 2. After the imputation, I ran the command sureg (seemingly unrelated regression).  However, Stata did not report the adjusted R-squared or RMSE.  Is there any way to get these statistics from the output?

If Stata protected you from referring to these quantities, there are
probably good reasons for doing so. The MI framework is only useful
for getting somewhat better point estimates and their standard errors,
and it is not appropriate for tests and measures of overall fit.

You would want to read more about MI and its limitations before
embarking on the project.

Stas Kolenikov, also found at
Small print: I use this email account for mailing lists only.

*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index