R help - Jul 2008 - Rolling range and regression calculations

Hi everyone,

I want to calculate a min and max (i.e. range) on a rolling time frame of 50
periods for a matrix with the number of periods extending along the row
direction. So for 300 periods, there will be 6 time frame windows per
column, and 6 min max pairs. I then want to o/p these pairs to a separate
matrix.

On a separate matter, I have a matrix containing data on which perform a
regresssion over a rolling time period. Is there a convenient way I can do
this for each column, and then save the gradient to an o/p matrix?

Thanks,

rcoder
-- 
View this message in context:
http://www.nabble.com/Rolling-range-and-regression-calculations-tp18643465p18643465.html
Sent from the R help mailing list archive at Nabble.com.

On Thu, 24 Jul 2008, rcoder wrote:
> Hi everyone,
>
> I want to calculate a min and max (i.e. range) on a rolling time frame of
50
> periods for a matrix with the number of periods extending along the row
> direction. So for 300 periods, there will be 6 time frame windows per
> column, and 6 min max pairs. I then want to o/p these pairs to a separate
> matrix.
>
> On a separate matter, I have a matrix containing data on which perform a
> regresssion over a rolling time period. Is there a convenient way I can do
> this for each column, and then save the gradient to an o/p matrix?
Look at the package "zoo", specifically the examples on the manual
pages
of ?rollapply and ?aggregate.zoo. These should be helpful in doing what 
you want. The package vignettes have further worked examples.

hth,
Z
> Thanks,
>
> rcoder
> -- 
> View this message in context:
http://www.nabble.com/Rolling-range-and-regression-calculations-tp18643465p18643465.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.
>
>

Hi Achim,

Thanks for your reply. rollmean and rollmax functions exist, but is there
anything for returning the minima on a rolling basis? I know there is no
rollmin in the zoo library.

Thanks,

rcoder



Achim Zeileis wrote:
> 
> On Thu, 24 Jul 2008, rcoder wrote:
> 
>> Hi everyone,
>>
>> I want to calculate a min and max (i.e. range) on a rolling time frame
of
>> 50
>> periods for a matrix with the number of periods extending along the row
>> direction. So for 300 periods, there will be 6 time frame windows per
>> column, and 6 min max pairs. I then want to o/p these pairs to a
separate
>> matrix.
>>
>> On a separate matter, I have a matrix containing data on which perform
a
>> regresssion over a rolling time period. Is there a convenient way I can
>> do
>> this for each column, and then save the gradient to an o/p matrix?
> 
> Look at the package "zoo", specifically the examples on the
manual pages
> of ?rollapply and ?aggregate.zoo. These should be helpful in doing what 
> you want. The package vignettes have further worked examples.
> 
> hth,
> Z
> 
>> Thanks,
>>
>> rcoder
>> -- 
>> View this message in context:
>>
http://www.nabble.com/Rolling-range-and-regression-calculations-tp18643465p18643465.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.
>>
>>
> 
> ______________________________________________
> 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.
> 
> 
-- 
View this message in context:
http://www.nabble.com/Rolling-range-and-regression-calculations-tp18643465p18650915.html
Sent from the R help mailing list archive at Nabble.com.