# Re: st: confidence intervals for locpoly

 From Maarten buis <[email protected]> To [email protected] Subject Re: st: confidence intervals for locpoly Date Mon, 25 Jan 2010 09:38:36 +0000 (GMT)

```--- On Mon, 25/1/10, Sharooon wrote:
> I can't get the bootstrap to work with locpoly.

If you are using a program that is not part of official Stata, then the
Statalist FAQ askes you to specify where you got. I am assuming you are
using the updated version that appeared in the Stata Journal 6(4) (,which
I found by typing -findit locpoly).

Below is an example on how to do that. However, you'll want to read up
on whether the simple bootstrap is appropriate in your case, for example
Chapters 8 and 9 of Efron and Tibshirani (1993). Still, even if it
doesn't produce meaningfull standard errors/confidence intervals, it can
be used to asses the stability of your estimates, as discussed by
Royston and Sauerbrei in the latest Stata Journal.

*----------------- begin example ------------------

sysuse auto, clear
locpoly mpg weight, nograph  ///
degree(3) width(500) ///
generate(xgrid yhat_obs)

keep if xgrid <.
keep xgrid yhat_obs
tempfile tofill
sort xgrid
save `tofill'
gen id = _n
keep id xgrid
sort id
tempfile grid
save `grid'

forvalues i = 1/500 {
qui {
sysuse auto, clear
bsample
gen id = _n
sort id
merge id using `grid'
drop _merge
locpoly mpg weight,      ///
degree(3) width(500) ///
generate(yhat`i')    ///
at(xgrid) nograph
keep if xgrid < .
keep xgrid yhat`i'
sort xgrid
merge xgrid using `tofill'
assert _merge == 3
drop _merge
sort xgrid
save `tofill', replace
}
_dots `i' 0
}
use `tofill'
egen lb = rowpctile(yhat? yhat?? yhat???), p(2.5)
egen ub = rowpctile(yhat? yhat?? yhat???), p(97.5)
twoway rarea lb ub xgrid || line yhat_obs xgrid
*------------------ end example ---------------------
( For more on how to use examples I sent to statalist see:
http://www.maartenbuis.nl/stata/exampleFAQ.html )

Hope this helps,
Maarten

Bradley Efron and Robert J. Tibshirani (1993) An Introduction to the
Bootstrap. Boca Raton: Chapman & Hall/CRC.

Patrick Royston and Willi Sauerbrei (2009) Bootstrap assessment of the
stability of multivariable models. The Stata Journal, 9(4): 547–570.
http://www.stata-journal.com/article.html?article=st0177

--------------------------
Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen
Germany

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

*
*   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/
```