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]
st: OLS and Heteroskedasticity
From
"Fernando Rios Avila" <[email protected]>
To
<[email protected]>
Subject
st: OLS and Heteroskedasticity
Date
Wed, 30 Mar 2011 19:11:30 -0400
Dear Stata listers
I was wondering if anyone knows how feasible is to use Oaxaca type of
decomposition, but considering not only wages but also dispersion?
I was planning to estimate models using ML function of stata and explicitly
include different controls for the estimation of the variance of the errors
in a mincer kind of equation.
Thanks in advance
-----Original Message-----
From: [email protected]
[mailto:[email protected]] On Behalf Of Austin Nichols
Sent: Wednesday, March 30, 2011 5:46 PM
To: [email protected]
Subject: Re: st: drawing lines to points outside the specified range of plot
You need an intermediate step calculating the intersections of your lines
with the boundaries of the plot region--in Mata, presumably.
Should take zero seconds to execute, approximately.
On Wed, Mar 30, 2011 at 3:11 PM, <[email protected]> wrote:
> Thanks a lot for the suggestions, however, I think I have not been
> clear enough in what my problem is so I'll try again;
>
> My goal is making plots like these:
>
> http://www.escholarship.org/uc/item/3b06127p?display=all#page-6
>
> (The paper gives R code which I'm trying to adapt for Mata)
>
> All the shown plots have (roughly) x-range(-2,2) y-range (-2,2.5) The
> red lines are drawn by connecting paired coordinates. However, some
> coordinates/points are way outside the plot region (for example
> (-100,100)), where the lines project towards, while at the same time
> constraining the plotsize/axis-range to the same size. However this I can
not manage in Stata:
> either the range has to be extended or the line is not drawn. So, as
> an simple
> example:
>
> twoway pci 1 1 30 30, xsc(r(0 50))
>
> can't be changed to
>
> twoway pci 1 1 30 30, xsc(r(0 5))
>
> since it then sets the range of the xsc to (0 30) automatically, which
> is normally the most logical of course.
>
> I hope this is a better description of the problem,
>
> regards Alex
*
* 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/