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: adjusted fitted-line/scatter plot after bootstrap with CI


From   Nick Cox <njcoxstata@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: adjusted fitted-line/scatter plot after bootstrap with CI
Date   Wed, 30 Nov 2011 08:56:46 +0000

Maarten's answer shows that this is far too pessimistic.

On Wed, Nov 30, 2011 at 8:42 AM, Nick Cox <njcoxstata@gmail.com> wrote:
> -qfitci- is _not_ for displays after model estimation. I think you are
> going to have to do this yourself.
> Confidence intervals can be shown using -rarea-.
>
> Nick
>
> On Wed, Nov 30, 2011 at 5:34 AM, Juan Dominguez
> <juan.dominguez@monash.edu> wrote:
>
>> I am fitting a model with a quadratic term (reg DV IV IV^2 COVs) and I
>> would like to graph the quadratic relationship after adjustment for
>> covariates and bootstrapping with confidence interval envelope much as
>> fracplot does after fracpoly.
>>
>> I tried cprplot but it won't work with bootstrap and won't add CI. In
>> addition, cprplot will draw a straight line rather than a curve
>> describing the quadratic relationship. The lowess option does not
>> yield a satisfactory curve like fracplot. acprplot will run after
>> bootstrapping (as in reg DV IV IV^2 COV, vce(boot, boot_options)), but
>> just like cprplot, it won't graph the quadratic curve and won't
>> include the CI envelope.
>>
>> Is there a way to accomplish this using qfitci?
>>
>> Best next thing is predxcon and while it does display a line fitted to
>> predicted values it can't be used in combination with bootstrap and it
>> won't include the associated scatter plot.
>>
>> In a previous posting on this list for a similar query it was
>> suggested to manually predict y and se after the adjusted regression,
>> and draw the regression line with ci's on top of the scatterplot. I
>> know how to predict y and se but I don't know how to draw the
>> subsequent (curved) regression line with ci's.
*
*   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/


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