Statalist The Stata Listserver

[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: RE: How does xtmixed/xtreg deal with missing data?

From   "Feiveson, Alan H. (JSC-SK311)" <>
To   <>
Subject   st: RE: How does xtmixed/xtreg deal with missing data?
Date   Tue, 15 May 2007 08:05:10 -0500

Sarah - 

Unfortunately it appears that all of Stata's regression commands or for
that matter any command that deals with a list of variables, will drop
observations that have at least one of the variables in the list
missing. So -xtmixed- is no exception. You can verify this by re-running
the analysis with incomplete observations deleted (by yourself) and
observe whether you get the same results. The alternative is to use one
of the multiple imputation methods (such as -ice-) to construact a
"complete" data set.

Al Feiveson

-----Original Message-----
[] On Behalf Of Sarah Gill
Sent: Monday, May 14, 2007 5:05 PM
Subject: st: How does xtmixed/xtreg deal with missing data?

I using xtmixed for longitudinal modelling of panel data. I know xtmixed
accommodates unbalanced datasets (i.e., some respondents drop out over
time or miss one or more waves of data collection), but what about
missing data within a wave/occasion, e.g., missing information on some,
but not all items within an occasion ? What happens in this instance?
Does all data for this respondent on this occasion get dropped similar
to listwise deletion?

Many thanks to any advice offered,

Sarah Gill
Centre for Mental Health Research
The Australian National University
Canberra ACT 0200 Australia
T: +61 2 6125 3379
F: +61 2 6125 0733

CRICOS Provider #00120C

*   For searches and help try:

*   For searches and help try:

© Copyright 1996–2015 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index