Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


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

Re: st: heteroskedasticity between groups, interpretation of -sdtest-


From   Maarten Buis <maartenlbuis@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: heteroskedasticity between groups, interpretation of -sdtest-
Date   Tue, 18 Dec 2012 19:53:34 +0100

I don't think anything got mixed up here. When you mean with
desriptive analysis a comparison of means, than that is equivalent to
running a regression with a indicator variable for survey. So, what
(slightly but not really) matters for inference is the difference in
variance of the residiuals not the difference in the raw variances.
People were kind and just assumed you and the referee knew that.

-- Maarten

On Tue, Dec 18, 2012 at 7:29 PM, Laura R. <laura.roh@googlemail.com> wrote:
> I'm sorry, I think something got mixed up here.
>
> I meant heterogeneity with reference to the comparison of variances
> between the two groups, i.e., the samples of the 2 surveys. I am not
> talking about the variance of residuals of the model (that includes
> both groups with a dummy for the survey year) here.
>
> It just seems that a referee seems concerned about making inferences
> from descriptive statistics on these two groups, if the variances are
> not equal.
>
> That is why I used -sdtest- to compare the variances between these
> groups. However, I was not sure how the information about whether
> variances significantly differ is important for the descriptive
> analysis or the regression, if a dummy is used there.
>
> Laura
>
>
>
> 2012/12/18 JVerkuilen (Gmail) <jvverkuilen@gmail.com>:
>> On Mon, Dec 17, 2012 at 6:12 AM, Nick Cox <njcoxstata@gmail.com> wrote:
>>> If your main interest is in the regression, the results of -sdtest-
>>> are at best marginal (pun intended) and at worst irrelevant.
>>
>> I'd argue it's even misleading, because the equality of variances test
>> is notorious for being non-robust.
>> *
>> *   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/
> *
> *   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/



-- 
---------------------------------
Maarten L. Buis
WZB
Reichpietschufer 50
10785 Berlin
Germany

http://www.maartenbuis.nl
---------------------------------
*
*   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/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index