Statalist


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

RE: st: Restricted Ranges in Stata's Multiple Imputation Procedures


From   "Hoogendoorn, Adriaan" <A.Hoogendoorn@ggzingeest.nl>
To   "'statalist@hsphsun2.harvard.edu'" <statalist@hsphsun2.harvard.edu>
Subject   RE: st: Restricted Ranges in Stata's Multiple Imputation Procedures
Date   Tue, 20 Oct 2009 11:14:27 +0200

Hey Maarten,

Some didactic and powerfull lines you wrote there!

Thank you for your suggestion of using the nbreg model as an option for the imputation model.
Also thanks a lot for pointing out the -stripplot- package by Nick Cox. A welcome tool to get diagnostics of the imputation process in an easy way.
The nbreg model proved a much better model than the standard OLS.
Although - as you already pointed out - it does not fix the problem at the upper bound, the main problem appeared to - as you suggested - at the lower bound and this was fixed by your solution.

*------------------ begin example ----------------------------
sysuse nlsw88, clear
// create some missing values
replace wage = . if runiform() < invlogit(-2 - union + south)

gen misswage = missing(wage)
ice wage union south grade, m(5) clear cmd(wage:nbreg)
stripplot wage if _mj > 0 , over(_mj) by(misswage) stack
*------------------- end example -----------------------------

Kind regards,
Adriaan Hoogendoorn
GGZ inGeest


*
*   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 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index