r - How to Repeat some consecutive lines Nth Times in R

Question

How can I repeat my line 22 to line 26 ten times serially and add the result of line 26 from first to the 10th and finally get the average of the 10 consecutive sum in my r code bellow:

# simulate arima(1,1,0)
library(forecast)
set.seed(100)
wn <- rnorm(10, mean = 0, sd = 1)
ts <- wn[1:2]
for (i in 3:10){
  ts<-arima.sim(n=10,model=list(ar=-0.7048,order=c(1,1,0)),start.innov=4.1,n.start=1,innov=wn)
}
ts <-ts[-1]
#
# write the function for RMSE
rmse <- function(x) {
  m <- auto.arima(x)
  acu <- accuracy(m)
  acu[1, 2]
}
#
t<-length(ts)# the length of the time series
l<- 2# block size
m <- ceiling(t / l) # number of blocks
blk<-split(ts, rep(1:m, each=l, length.out = t)) # divides the series into blocks
res<-sample(blk, replace=T, 1000) # resamples the blocks
unlist<-unlist(res, use.names = F) # unlist the bootstrap series
tsunlist<-ts(unlist) # turns the bootstrap series into time series data
# use the RMSE function
RMSE <- rmse(tsunlist)

The above R code performs the following algorithm in steps:

1. Simulate ARIMA(1,1,0) time series (line 1 to line 9)
2. Split the time series into blocks of equal size of 2 (line 18 to line 21)
3. Resample each block at random 1000 times (line 22 to line 23)
4. Rearrange the resampled series into time series data (line 24)
5. Obtain the RMSE of the resampled time series (line 26)

I want to repeat step 3 to 5 100 times, add up the results produced in step 5 100 times and get the average of the results.

标签： rloopsrep

Answer 1

Here is an approach for how to perform the resample 100 times. Replace your code starting on line 22 with this.

finalresult <- vector() #initialize vector to receive result from for loop
for(i in 1:100){
   res<-sample(blk, replace=T, 1000) # resamples the blocks
   res.unlist<-unlist(res, use.names = F) # unlist the bootstrap series
   tsunlist<-ts(res.unlist) # turns the bootstrap series into time series data
   # use the RMSE function
   RMSE <- rmse(tsunlist)
   finalresult[i] <- RMSE # Assign RMSE value to final result vector element i
}

Once you run the loop you will find the results in the finalresult object.

finalresult
  [1] 0.4695763 0.4702997 0.4727734 0.4677841 0.4633566 0.4619570 0.4645237 0.4657333 0.4734756 0.6035923 0.4676718 0.4563636
 [13] 0.4653432 0.4741868 0.4638185 0.4679926 0.4652872 0.4644442 0.4673774 0.4654423 0.4574012 0.4613827 0.4689168 0.4652262
 [25] 0.4621680 0.4714052 0.4544405 0.4781833 0.4640436 0.4708187 0.4562165 0.4631720 0.4638906 0.4654569 0.4691919 0.4644442
 [37] 0.4635361 0.4657427 0.4682825 0.4626979 0.4636363 0.4562305 0.4582826 0.4689343 0.4648181 0.4659912 0.4597617 0.4701386
 [49] 0.4678786 0.4658028 0.4633929 0.4759015 0.4719038 0.4685480 0.4639831 0.4663984 0.4631158 0.4636240 0.4677248 0.4680744
 [61] 0.4633999 0.4734937 0.4545364 0.4671157 0.4656818 0.4638791 0.4676848 0.4650673 0.4710859 0.4680129 0.4641621 0.4632793
 [73] 0.4664942 0.4596942 0.4643477 0.4626547 0.4684443 0.4572568 0.4695198 0.4566412 0.4650888 0.4643392 0.4603766 0.4694628
 [85] 0.4579581 0.4627443 0.4729837 0.4668802 0.4723182 0.4688657 0.4684541 0.4655471 0.4687285 0.4504862 0.4599657 0.4660643
 [97] 0.5093069 0.4620558 0.4655066 0.4682742