Re: st: mi impute chained error messages

["JVerkuilen (Gmail)" <jvverkuilen@gmail.com>]

On Tue, Oct 23, 2012 at 2:26 AM, chong shiauyun <shiauyun416@hotmail.com> wrote:
> Thanks for your response. I will try linear regression or I may use truncated regression that allows me to specify the lower and the upper limit. Actually I did manage to run the dryrun when I put my syntax like this
> mi impute chained (reg) .... (ologit)... (intreg, ll(lTIQ) ul(uTIQ))TIQ, add(20),... dryrun report (just have to put the lower and upper limit in the same blanket as the (intreg) and then specify a new variable, TIQ, that have the impputed value of IQ.
>
> I think truncated regression would be more suitable to apply to my IQ scores (which are continous variables with boundaries between 45-155).I wil also try linear regression.

I suspect you won't have too many near the boundary so don't sweat it.
It makes the imputation more complicated.

Also, don't overlook predictive mean matching (pmm) as a method when
you have boundaries and such, as long as the N is large.



> Another thing that I would like to know is that does it matter if I put variables that I have collapsed (eg. for education level, I collapsed bachelor degree and postgraduate degree into a group called high education) instead of the original data into the imputation model?

Ideally the imputation model mirrors the behavior in the data as well
as possible. This means that it's often more complex than the intended
analysis model, to some degree, and may include things like nonlinear
terms for a given variable when you plan on using a linear term in the
model, or extra interactions. Dirty cases or variables have their
influence spread all over the dataset, though, just like in any other
full information method, so be wary.

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/