[R-SIG-Finance] rollapply over several columns

[Ulrich Staudinger]

Yes,
that's it indeed. I didn't read the documentation precisely enough and
assumed it would be FALSE by default (which I find more intuitive).
Cheers

On Thu, Sep 9, 2010 at 11:54 AM,  <Murali.Menon at avivainvestors.com> wrote:
> The by.column = FALSE argument in rollapply() should help?
> Murali
>
> -----Original Message-----
> From: r-sig-finance-bounces at stat.math.ethz.ch [mailto:r-sig-finance-bounces at stat.math.ethz.ch] On Behalf Of Ulrich Staudinger
> Sent: 09 September 2010 10:10
> To: r-sig-finance
> Subject: [R-SIG-Finance] rollapply over several columns
>
> Hi list,
>
>
> I am trying to use rollapply on am wondering about the following question:
> 1) How can I tell rollapply to process my data not column after column but all columns at once?
>
>
> Consider the following data set (a record of quotes):
>
>> part
>                         V2 V3     V4 V5
> 2010-08-22 23:03:18 105500 12 107400  1
> 2010-08-22 23:03:18 105500 12 107400  1
> 2010-08-22 23:03:18 105500 12 107400  1
> 2010-08-22 23:03:18 105500 12 107400  1
> 2010-08-22 23:03:18 105500 12 107400  1
> 2010-08-22 23:03:18 105500 12 107400  1
> 2010-08-22 23:03:18 105500 12 107400  1
> 2010-08-22 23:03:18 105500 12 107400  1
> 2010-08-22 23:03:18 105500 12 107400  1
> 2010-08-22 23:03:18 105500 12 107400  1
>> class(part)
> [1] "xts" "zoo"
>>
>
> ... and the following very simply function ....
>
>
> testFunction <- function(x)
> {
>     cat(class(x), "///", x, "\n")
>     rnorm(1)
> }
>
>
> .... the output of using rollapply on this is:
>
>
>> rollapply(part, 4, testFunction)
> numeric /// 105500 105500 105500 105500
> numeric /// 105500 105500 105500 105500
> numeric /// 105500 105500 105500 105500
> numeric /// 105500 105500 105500 105500
> numeric /// 105500 105500 105500 105500
> numeric /// 105500 105500 105500 105500
> numeric /// 105500 105500 105500 105500
> numeric /// 12 12 12 12
> numeric /// 12 12 12 12
> numeric /// 12 12 12 12
> numeric /// 12 12 12 12
> numeric /// 12 12 12 12
> numeric /// 12 12 12 12
> numeric /// 12 12 12 12
> numeric /// 107400 107400 107400 107400
> numeric /// 107400 107400 107400 107400
> numeric /// 107400 107400 107400 107400
> numeric /// 107400 107400 107400 107400
> numeric /// 107400 107400 107400 107400
> numeric /// 107400 107400 107400 107400
> numeric /// 107400 107400 107400 107400
> numeric /// 1 1 1 1
> numeric /// 1 1 1 1
> numeric /// 1 1 1 1
> numeric /// 1 1 1 1
> numeric /// 1 1 1 1
> numeric /// 1 1 1 1
> numeric /// 1 1 1 1
>                              V2         V3         V4         V5
> 2010-08-22 23:03:18 -0.26829508  1.3107688  1.4787251 -0.2583725
> 2010-08-22 23:03:18 -2.17382434  0.1600959  0.4368223 -1.3948148
> 2010-08-22 23:03:18 -0.48042277 -0.3413495 -0.8816699 -0.2819355
> 2010-08-22 23:03:18  0.88840373 -0.7273423 -1.1688194  0.2605899
> 2010-08-22 23:03:18 -0.04364438 -0.3462516 -1.2633264 -0.7175950
> 2010-08-22 23:03:18 -0.79960171  0.2439625  0.1893828 -1.0448513
> 2010-08-22 23:03:18  0.53943691  1.2008782  0.2796326 -1.4020305
>>
>
> ......... as you can see, it processes column after column, with a nice rolling window. But I would like it to hand in all four columns at once, with a sliding window.
>
> How?
>
> Thanks in advance,
> Ulrich
>
>
>
> --
> Ulrich Staudinger
> http://new.activequant.org
>
> _______________________________________________
> R-SIG-Finance at stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-sig-finance
> -- Subscriber-posting only. If you want to post, subscribe first.
> -- Also note that this is not the r-help list where general R questions should go.



-- 
Ulrich Staudinger
Klosterstr. 7
CH-6003 Luzern
Switzerland
++41-79-7020595