R help - Jun 2010 - Fast and simple tool for re-sampling of asynchronous time series ?

Un texte encapsul? et encod? dans un jeu de caract?res inconnu a ?t? nettoy?...
Nom : non disponible
URL :
<https://stat.ethz.ch/pipermail/r-help/attachments/20100625/44fbea79/attachment.pl>

On Fri, 25 Jun 2010, bruno Piguet wrote:
> Hi all,
>
>   I'm looking for a function which could do some fast and simple
> re-sampling of asynchronous time series.
>
>   Below is a MCE of the kind of algorithm I need. As you can see, it's
> quite crude, but it's enough for my current needs.  The only problem is
that
> it is quite slow on real use case.
>   I've got a C version which is much faster, but I'd like to have a
pure-R
> program.
>
>   Any pointer to the relevant part of the doc one one of the time-series
> packages ? Any suggestion or advice ?
>
>   Thanks in advance,
>
> B. Piguet.
>
> Here is the exemple :
> Tx <- seq(1, 50, 0.5)
> Tx <- Tx + rnorm(length(Tx), 0, 0.1)
> X <- sin(Tx/10.0) +  sin(Tx/5.0) + rnorm(length(Tx), 0, 0.1)
> Ty <- seq(1, 50, 0.3333)
> Ty <- Ty + rnorm(length(Ty), 0, 0.02)
> Y <- sin(Ty/10.0) + sin(Ty/5.0) + rnorm(length(Ty), 0, 0.1)
>
> w <- 0.25

Personally, I'd incline towards leaving the next lines to C, perhaps using 
the inline package.

But if you want a purely R solution, the bioConductor IRanges package 
should help. I think the viewMeans() function will handle this loop.

See

 	http://comments.gmane.org/gmane.comp.lang.r.sequencing/1296

for some discussion.

HTH,

Chuck
>
> Y_sync <- rep(NA, length(Tx))
> for (i in 1:length(Tx))
> {
>   T_min <- Tx[i] - w
>   T_max <- Tx[i] + w
>   Y_sync[i] <- mean(Y[Ty >= T_min & Ty <= T_max ])
> }
>
> diff = X - Y_sync
> print(summary(diff))
>
> print(summary(lm(Y_sync~X)))
> plot (diff~Tx, type="l")
>
> 	[[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.
>
Charles C. Berry                            (858) 534-2098
                                             Dept of Family/Preventive Medicine
E mailto:cberry at tajo.ucsd.edu	            UC San Diego
http://famprevmed.ucsd.edu/faculty/cberry/  La Jolla, San Diego 92093-0901

On Fri, Jun 25, 2010 at 11:34 AM, bruno Piguet <bruno.piguet at gmail.com>
wrote:
> Hi all,
>
> ? I'm looking for a function which could do some fast and simple
> re-sampling of asynchronous time series.
>
> ? Below is a MCE of the kind of algorithm I need. As you can see, it's
> quite crude, but it's enough for my current needs. ?The only problem is
that
> it is quite slow on real use case.
> ? I've got a C version which is much faster, but I'd like to have a
pure-R
> program.
>
> ? Any pointer to the relevant part of the doc one one of the time-series
> packages ? Any suggestion or advice ?
>
> ? Thanks in advance,
>
> B. Piguet.
>
> Here is the exemple :
> Tx <- seq(1, 50, 0.5)
> Tx <- Tx + rnorm(length(Tx), 0, 0.1)
> X <- sin(Tx/10.0) + ?sin(Tx/5.0) + rnorm(length(Tx), 0, 0.1)
> Ty <- seq(1, 50, 0.3333)
> Ty <- Ty + rnorm(length(Ty), 0, 0.02)
> Y <- sin(Ty/10.0) + sin(Ty/5.0) + rnorm(length(Ty), 0, 0.1)
>
> w <- 0.25
>
> Y_sync <- rep(NA, length(Tx))
> for (i in 1:length(Tx))
> {
> ? T_min <- Tx[i] - w
> ? T_max <- Tx[i] + w
> ? Y_sync[i] <- mean(Y[Ty >= T_min & Ty <= T_max ])
> }
>
> diff = X - Y_sync
> print(summary(diff))
>
> print(summary(lm(Y_sync~X)))
> plot (diff~Tx, type="l")
This isn't substantially different than what you have but does replace
the explicit loop and associated indexing with an implicit loop:

   sapply(Tx, function(tx) mean(Y[Ty >= tx-w & Ty <= tx+w]))

Un texte encapsul? et encod? dans un jeu de caract?res inconnu a ?t? nettoy?...
Nom : non disponible
URL :
<https://stat.ethz.ch/pipermail/r-help/attachments/20100629/335df200/attachment.pl>