Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.

# Re: st: Lpoly of an estimated/computed variable

 From Derya Karaci To "statalist@hsphsun2.harvard.edu" Subject Re: st: Lpoly of an estimated/computed variable Date Thu, 10 Jan 2013 06:49:57 -0800 (PST)

```Hi Austin. Thanks for your reply. I would like to smooth Y only... I am estimating a price effect for each household, and would like to smooth it across income distribution. X,Z,K are different components of the price effect (none of them are income). I can assume covariances as zero...

My main question is how use lpoly on a variable that is estimated and has associated standard errors. Any help will be greatly appreciated.

Regards

----- Original Message -----
From: Austin Nichols <austinnichols@gmail.com>
To: statalist@hsphsun2.harvard.edu
Cc:
Sent: Wednesday, January 9, 2013 7:18 PM
Subject: Re: st: Lpoly of an estimated/computed variable

Derya Karaci <dkaraci@yahoo.com>:
Across what values of what variable are you smoothing? X? Z? Both?
Something else?
I can't see how you could use the SEs on coefs unless you also have
the covariances.

On Wed, Jan 9, 2013 at 7:14 PM, Derya Karaci <dkaraci@yahoo.com> wrote:
> Hi all,
>
> I need to calculate a new variable using a formula involving
> other variables and coefficients. Y=aX+bZ+cK, etc. The coefficients a, b and c
> are estimated elsewhere and each have standard errors. Secondly, I would like
> to estimate a local polynomial smooth of Y where confidence bands also incorporate
> the standard errors of a, b and c.
>
> How can I do this? I would be most grateful if you can give
> me an idea. I can give more details...
*
*   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/
```