[R] Moving-Tiles Bootstrap

R. Michael Weylandt michael.weylandt at gmail.com
Wed Jan 25 02:02:52 CET 2012 

So I'll preface this with the admission I have no idea what
moving-tiles or boot.tilt are and so I may be in way over my head, but
if you are having trouble fitting your bootstrap procedure into one of
those provided in package, it's not hard to roll your own with
replicate() in base R.

The key is going to be clarifying "I'm not simply resampling rows of a
data frame" -- how exactly are you resampling? If you can wrap it in a
function, you can pass it to replicate.

Michael

PS -- If these are all domain-specific techniques, you might check the
SIG groups: I've found the ones that I participate in to be full of
exceptionally knowledgeable help.

On Tue, Jan 24, 2012 at 7:41 PM, Jeff Newmiller
<jdnewmil at dcn.davis.ca.us> wrote:
> I wouldn't characterize the requirement to provide R code as simply "nuts and bolts", but rather a preference for precision in communication. Very definitely there are theoretical discussions here, but for such discussions short R examples are used to make sure the concepts are understood.
> ---------------------------------------------------------------------------
> Jeff Newmiller                        The     .....       .....  Go Live...
> DCN:<jdnewmil at dcn.davis.ca.us>        Basics: ##.#.       ##.#.  Live Go...
>                                      Live:   OO#.. Dead: OO#..  Playing
> Research Engineer (Solar/Batteries            O.O#.       #.O#.  with
> /Software/Embedded Controllers)               .OO#.       .OO#.  rocks...1k
> ---------------------------------------------------------------------------
> Sent from my phone. Please excuse my brevity.
>
> Kim Elmore <kim.elmore at noaa.gov> wrote:
>
>>I wish to proffer my sincere apologies to both you and the list. I had
>>not seen my posting appear -- it was initially held for moderator
>>approval, but I never saw that it was released. During the subscription
>>
>>process, I had used an incorrect e-mail address and so withdrew the
>>post. Within my subscription settings, I had asked to see my own
>>postings but, so far have not. I have now doubled checked that this is
>>the case. Hence, the multiple posts. Again, apologies. Finally, this
>>seems to be a more nuts-and-bolts list and that is not intended to
>>address the admittedly esoteric topic about which I posted my query so,
>>
>>again, apologies to all. Finally, I think I have set Thunderbird to
>>strip off all HTML in this posting. Again, apologies for the HTML in
>>prior posts -- I certainly intend for there to be none in this one.
>>
>>Kim Elmore
>>
>>On 1/24/2012 11:31 AM, Jeff Newmiller wrote:
>>> A) Reposting the same message as a separate thread is bad netiquette.
>>>
>>> B) Posting HTML email on this list is strongly discouraged.
>>>
>>> C) You probably ought to read the posting guide.
>>>
>>> D) This is a list about using R. Posting theoretical questions
>>without corresponding R code that illustrates what how far you have
>>progressed is unlikely to elicit a response.
>>>
>>> E) Some questions require just the right combination of expertise and
>>free time to answer. Yours seems to be bleeding into a theoretical
>>discussion of a very specific knowledge domain... I certainly don't
>>know the answer. You may need to correspond with authors of packages
>>that almost apply to your problem or post on a more focused discussion
>>group (perhaps one that discusses statistical theory) if you don't get
>>an answer here. Or just wait a week for the right person to read your
>>question.
>>>
>>---------------------------------------------------------------------------
>>> Jeff Newmiller                        The     .....       .....  Go
>>Live...
>>> DCN:<jdnewmil at dcn.davis.ca.us>         Basics: ##.#.       ##.#.
>>Live Go...
>>>                                        Live:   OO#.. Dead: OO#..
>>Playing
>>> Research Engineer (Solar/Batteries            O.O#.       #.O#.  with
>>> /Software/Embedded Controllers)               .OO#.       .OO#.
>>rocks...1k
>>>
>>---------------------------------------------------------------------------
>>> Sent from my phone. Please excuse my brevity.
>>>
>>> Kim Elmore<kim.elmore at noaa.gov>  wrote:
>>>
>>>> I wish to perform moving tiles bootstrap resampling on some gridded
>>>> data
>>>> meteorological data. I've many years experience with S-Plus, but it
>>has
>>>>
>>>> no way to perform a moving-tiles bootstrap. Within R I've learned
>>how
>>>> to
>>>> use quadratresample() with the spatstats package and would be happy
>>to
>>>> simply use empirical percentiles if generating the replicates were
>>>> fast,
>>>> but it isn't. So, I'd like to employ bootstrap tilting to generate
>>my
>>>> confidence intervals. On my machine (older Athlon, XP sp3, 4 GB) it
>>>> took
>>>> about 48 h to generate 5000 moving tiles replicates. But, the boot
>>>> package contains boot.tilt and I should be able to do well enough
>>with
>>>> the statistics I need (mean and RMS) to get by with several hundred
>>>> samples instead of several thousand.
>>>>
>>>> I can certainly build a function that generates the statistic I want
>>> >from a moving-tiles replicate. My problem is that the resampling
>>>> process
>>>> isn't as simple as boot.tilt() expects (I'm not simply resampling
>>rows
>>>> of a data frame). But, the function that does what I need doesn't
>>fit
>>>> into how boot.tilt() expects the resampling process to work. I can
>>>> generate a string of replicates easily enough -- how might I wrap
>>such
>>>> a
>>>> set of replicates into an object on which tilting could be performed
>>>> absent the internal generation of the replicates themselves?
>>>>
>>>> Kim Elmore
>>>> --
>>>>
>>>> Kim Elmore, Ph.D. (CCM, PP SEL/MEL/Glider, N5OP, 2nd Class
>>>> Radiotelegraph, GROL)
>>>>
>>>> /"Human beings, who are almost unique in having the ability to learn
>>> >from the experience/
>>>> /of//others, are also remarkable for their apparent disinclination
>>to
>>>> do
>>>> so/."/-- Douglas Adams/
>>>>
>>>>
>>>>     [[alternative HTML version deleted]]
>>>>
>>>> ______________________________________________
>>>> 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.
>>//
>
> ______________________________________________
> 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.

More information about the R-help mailing list