Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
<<winmail.dat>>
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
RE: st: LOWESS: confidence band
From
<[email protected]>
To
<[email protected]>
Subject
RE: st: LOWESS: confidence band
Date
Fri, 2 Mar 2012 18:41:21 -0000
Dear Austin,
thank you very much for your advice.
Do you have any reference so I can better ground your suggestions?
(i.e. why triangle kernel and why degree 1?)
Many thanks,
fadi
________________________________
From: [email protected] on behalf of Austin Nichols
Sent: Fri 02/03/2012 17:12
To: [email protected]
Subject: Re: st: LOWESS: confidence band
fadi <[email protected]>:
Also use option degree(1) which should be the default, not degree(0).
On Fri, Mar 2, 2012 at 12:10 PM, Austin Nichols <[email protected]> wrote:
> fadi <[email protected]>:
> Conf bands at the boundary are tricky at best; try -lpoly- with a
> triangle kernel.
>
> On Fri, Mar 2, 2012 at 11:20 AM, <[email protected]> wrote:
>> do you know how to construct a 95% confidence band for a LOWESS estimation not by bootstrapping?
>> (I have a limited sample and I am interested in the confidence bands at boundary)
>> fadi
*
* 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/
Please access the attached hyperlink for an important electronic communications disclaimer: http://lse.ac.uk/emailDisclaimer