[R] Implementing Silverman's Negative Reflection

David Winsemius dwinsemius at comcast.net
Fri Sep 30 17:13:41 CEST 2011


On Sep 30, 2011, at 10:14 AM, Duncan Murdoch wrote:

> On 30/09/2011 10:00 AM, David Winsemius wrote:
>> On Sep 30, 2011, at 4:50 AM, sreblam wrote:
>>
>> >  Hi!
>> >  I'm trying to implement in R in an easy way the negative  
>> reflection
>> >  method
>> >  described in Silverman (1986) on p.31, ie I have a non-negative
>> >  dataset and
>> >  would like to estimate the density by applying a reflection method
>> >  where the
>> >  reflected points have weight -1.
>> >  I thought there should be a way to evaluate the wanted kernel at  
>> the
>> >  required points, in other words to calculate formula (2.16):
>> >  fhat(x) = 1/(nh) * sum_{i=1}^{n}{ K( (x-Xi) /h) - K( (x-Xi) /h) }
>> >
>> >  Unfortunately, I am still not that good and skilled with R - does
>> >  anybody
>> >  know how to do this- Or maybe a hint...?
>> >  Thank you!!
>>
>> Doing what I thought was the obvious search ( "negative reflection
>> density" )  done at the obvious search site (the one to which
>> RSiteSearch("negative reflection density") ) typed at the console
>> would have taken you) and twiddling with the setting produces this
>> perfect match as its second hit:
>>
>> http://finzi.psych.upenn.edu/Rhelp10/2010-July/246526.html
>
> That's a reflection density estimate, but not a "negative  
> reflection" estimate.  The latter aims for zero density at location  
> 0, not zero gradient to the density there.

PLease ignore my first response. I see that the problem is not solved  
as easily as I fantasied. The logspline package is given by otehrs in  
the same thread as offering a mechanism for setting boundaries on  
densities.

-- 
David.
>
> Duncan Murdoch
>
>> Now it is true that you wouldn't have gotten that with the console
>> entered string which is why I created a new function (and a new
>> browser button that fiddles with the default setting for RSiteSearch:
>>
>> rhelpSearch<- function(string,
>>                    restrict = c("Rhelp10", "Rhelp08", "Rhelp02",
>> "functions" ),
>>                    matchesPerPage = 100, ...)
>>           RSiteSearch(string=string,  restrict = restrict,
>> matchesPerPage = matchesPerPage, ...)
>>
>> R-search button mapped to:
>> http://search.r-project.org/cgi-bin/namazu.cgi?query=&max=100&result=normal&sort=score&idxname=Rhelp10&idxname=Rhelp08&idxname=Rhelp02&idxname=functions
>>
>>
>> >
>> >  PS: you can also find the mentioned passage on p.21 of
>> >  http://ned.ipac.caltech.edu/level5/March02/Silverman/paper.pdf
>> >
>> >  --
>> >  View this message in context: http://r.789695.n4.nabble.com/Implementing-Silverman-s-Negative-Reflection-tp3859191p3859191.html
>> >  Sent from the R help mailing list archive at Nabble.com.
>> >
>> >  ______________________________________________
>> >  R-help at r-project.org mailing list
>> >  https://stat.ethz.ch/mailman/listinfo/r-help
>> >  PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
>> >  and provide commented, minimal, self-contained, reproducible code.
>>
>> David Winsemius, MD
>> West Hartford, CT
>>
>> ______________________________________________
>> R-help at r-project.org mailing list
>> https://stat.ethz.ch/mailman/listinfo/r-help
>> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
>> and provide commented, minimal, self-contained, reproducible code.
>

David Winsemius, MD
West Hartford, CT



More information about the R-help mailing list