R help - Dec 2010 - Replacing the for loop for time series buid-up

Hi,
is there a function that replaces the following code?

n=200
boot.x[1]=odhad+boot.res[1]  #(boot.x[0]=1)

      for (j in 1:(n-1)) {
      boot.x[j+1]=odhad*boot.x[j]+boot.res[j+1]
      }

This is nested in two other loops, and I am looking for some way to improve
code performance
I tried sapply and cumprod but no success.

Thanks
Jan
-- 
View this message in context:
http://r.789695.n4.nabble.com/Replacing-the-for-loop-for-time-series-buid-up-tp3094421p3094421.html
Sent from the R help mailing list archive at Nabble.com.

On Dec 19, 2010, at 8:08 AM, Torch wrote:
>
> Hi,
> is there a function that replaces the following code?
>
> n=200
> boot.x[1]=odhad+boot.res[1]  #(boot.x[0]=1)
No. there is no boot.x[0] ... in R anyway.
>
>      for (j in 1:(n-1)) {
>      boot.x[j+1]=odhad*boot.x[j]+boot.res[j+1]
>      }
>
> This is nested in two other loops, and I am looking for some way to  
> improve
> code performance
> I tried sapply and cumprod but no success.
Probably because you didn't use paper and pencil to figure out what  
x_n was going to be.

You end up with a series that looks like:

boot.x[n] = odhad^n + boot.res[1]*odhad^(n-1) +boot.res[2]*odhad^(n-2)  
+ ... boot.res[n]

I'm thinking you should play around with n=20  instead of 200 because  
I don't know what odhad might be and I'd be afraid of overflow with  
the ^200 operation. Here's a failed attempt:

geo_od <- odhad^(19:0)
geo_res <- boot.res*geo_od
boot.x <- odhad^(1:20) + cumsum(geo_res)

But that's not correct ... except for the fact that it does get the  
last term correct.
>
> Thanks
> Jan
> -- 
> View this message in context:
http://r.789695.n4.nabble.com/Replacing-the-for-loop-for-time-series-buid-up-tp3094421p3094421.html
> Sent from the R help mailing list archive at Nabble.com.
David Winsemius, MD
West Hartford, CT

Torch wrote:
> 
> Hi,
> is there a function that replaces the following code?
> 
> n=200
> boot.x[1]=odhad+boot.res[1]  #(boot.x[0]=1)
> 
>       for (j in 1:(n-1)) {
>       boot.x[j+1]=odhad*boot.x[j]+boot.res[j+1]
>       }
> 
> This is nested in two other loops, and I am looking for some way to
> improve code performance
> I tried sapply and cumprod but no success.
> 
You can have a look at filter.
Test example:

alpha <- 0.75
N <- 10

boot.res <- ts(rnorm(N))
boot.res

val.init <- boot.res[1] + alpha

boot.x <- ts(val.init, start=1,end=N)
for(t in 2:N) { boot.x[t] <- alpha*boot.x[t-1] + boot.res[t] }
boot.x

boot.x.filter <-
ts(c(val.init,filter(boot.res[-1],filter=alpha,method="recursive",init=val.init)))
boot.x.filter
boot.x - boot.x.filter

succes

Berend


-- 
View this message in context:
http://r.789695.n4.nabble.com/Replacing-the-for-loop-for-time-series-buid-up-tp3094421p3094757.html
Sent from the R help mailing list archive at Nabble.com.