Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: st: HLM and SUR
From
Nick Cox <[email protected]>
To
[email protected]
Subject
Re: st: HLM and SUR
Date
Mon, 26 Nov 2012 13:25:40 +0000
If you think that -xtmixed- is the appropriate modelling command,
linear regression can hardly be a good substitute.
Conversely, if this makes sense to you why not try it?
Nick
On Mon, Nov 26, 2012 at 12:31 PM, KOTa <[email protected]> wrote:
> thanks. is it the only way?
>
> in my case it's not really feasible, as i have both lots and various
> number of observations within each cluster/group by which i split data
> for HLM. so reshaping into wide is problematic.
>
> how problematic (in terms of using the results of such analysis) would
> be to use .sureg and simple linear regression instead of xtmixed for
> such data?
> i mean would those results be "correct" ?
> On Mon, Nov 26, 2012 at 2:09 PM, Maarten Buis <[email protected]> wrote:
>> On Mon, Nov 26, 2012 at 12:36 PM, Constantin Alba wrote:
>>> is it possible to use SUR (seemingly unrelated regression) with HLM models?
>>>
>>> i have to 2 HLM models that i run separately where independent variables
>>> are same, and DV are different. Is there a way to run SUR while keeping HLM
>>> models intact?
>>
>> Both SUR and HLM can be seen as special cases of a SEM (see:
>> <http://blog.stata.com/2011/09/28/multilevel-random-effects-in-xtmixed-and-sem-the-long-and-wide-of-it/>),
>> so you can probably, with quite a bit of creative data preparation and
>> modeling, estimate such a model using -sem-.
*
* 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/