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: RE: RE: Test for Heteroskedasticity
From
Nick Cox <[email protected]>
To
"'[email protected]'" <[email protected]>
Subject
RE: st: RE: RE: Test for Heteroskedasticity
Date
Tue, 24 Jan 2012 18:43:25 +0000
There's no point of English at stake here.
The help for -whitetst- explains that it is intended for use after -regress- or -cnsreg-. It makes no sense to try to use it after -tobit-.
Looking at the others, -clad- (STB) also seems completely unrelated to -tobit-.
The more important question is: If there is a standard test for heteroscedasticity after -tobit-, why didn't StataCorp provide it?
Nick
[email protected]
Pattarawan Watcharaanantapong
Sorry if I use the wrong word becaue I am not a native english speaker
but I do not how to explain. My question looks silly for somebody but
I have no strong background about using stata, I just start to use
Tobit model, so I have no idea about it. Maybe, I might have too many
variables. Thank you for Tony.
Sorry Nick, if you use inappropriate word to you.
Tan
On Tue, Jan 24, 2012 at 11:39 AM, Nick Cox <[email protected]> wrote:
> I'm with Tony here in the bemused corner of the stand.
>
> Quite what would the test be? Can you point to literature on an appropriate test? Even the idea of a residual seems problematic for a tobit model.
>
> Also, I am down as an author of -whitetst- (STB/SSC): the help says explicitly "-whitetst- is for use after -regress- or -cnsreg-", so invoking it is pointless and comment that it "does not work" is inappropriate.
>
> Nick
> [email protected]
>
> Lachenbruch, Peter [Tony]
>
> A couple of comments:
> 1. Since Tobit seems to be working with truncated normals, doesn't this guarantee unequal variance?
> 2. Can you interpret a regression model with 35 predictors? I can't even in multiple regression.
>
> Pattarawan Watcharaanantapong
>
> I need to test for Heteroskedasticity after I run Tobit model,
> do anyone know about stata command to test for heteroskedaticity for
> tobit model? I tried whitetst, hettest, and clad but they did not work
> after running Tobit.
>
> For example: I ran this regression below
>
> tobit y x1 x2 x3 x4 x5 x6 x7 x8 x9 x10 x11....x35 [pweight], ll(0)
> whitetst
> last estimates not found
>
> Therefore, I am not sure I use the right command to test for
> Heteroskedasticity or not. Thank you so much.
>
> *
> * 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/
*
* 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/
*
* 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/