Timeseries Crossvalidation in R: using tsCV() with tslm()

Machine Learning and Modeling
, ,
1

Hi eveyrone

I am trying to evaluate a tslm-model using time series cross validation I am using the very simple code below (based on what Rob Hyndman posted on here some years ago), but I only get NAs as a result. What is the issue with the code? I must be missing something on how the tscv() function works

y = ts(rnorm(336),frequency=12)

fc <- function(y, h, xreg)
{
  if(NROW(xreg) < length(y) + h)
    stop("Not enough xreg data for forecasting")
  X <- xreg[seq_along(y),]
  fit <- tslm(y ~ X)
  X <- xreg[length(y)+seq(h),]
  forecast(fit, newdata=X)
}

 
pred <- ts(rnorm(length(y)), start=start(y),
           frequency=frequency(y))


tsCV(y, fc, xreg=pred)

  Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
1   NA  NA  NA  NA  NA  NA  NA  NA  NA  NA  NA  NA
2   NA  NA  NA  NA  NA  NA  NA  NA  NA  NA  NA  NA
3   NA  NA  NA  NA  NA  NA  NA  NA  NA  NA  NA  NA
4   NA  NA  NA  NA  NA  NA  NA  NA  NA  NA  NA  NA
5   NA  NA  NA  NA  NA  NA  NA  NA  NA  NA  NA  NA
6   NA  NA  NA  NA  NA  NA  NA  NA  NA  NA  NA  NA
7   NA  NA  NA  NA  NA  NA  NA  NA  NA  NA  NA  NA
8   NA  NA  NA  NA  NA  NA  NA  NA  NA  NA  NA  NA
9   NA  NA  NA  NA  NA  NA  NA  NA  NA  NA  NA  NA
10  NA  NA  NA  NA  NA  NA  NA  NA  NA  NA  NA  NA
11  NA  NA  NA  NA  NA  NA  NA  NA  NA  NA  NA  NA
12  NA  NA  NA  NA  NA  NA  NA  NA  NA  NA  NA  NA
13  NA  NA  NA  NA  NA  NA  NA  NA  NA  NA  NA  NA
14  NA  NA  NA  NA  NA  NA  NA  NA  NA  NA  NA  NA
2

Please read the help file for tsCV(). If you use exogenous regressors, the forecastfunction must have xreg and newxreg arguments corresponding to the training and test periods. There is an example provided in the help.

library(forecast)
set.seed(1)
options(digits = 2)

y <- ts(rnorm(336), frequency = 12)

fc <- function(y, h, xreg, newxreg) {
  fit <- tslm(y ~ xreg)
  forecast(fit, newdata = newxreg)
}

pred <- ts(rnorm(length(y)), start = start(y), frequency = frequency(y))

tsCV(y, fc, xreg = pred)
#>        Jan     Feb     Mar     Apr     May     Jun     Jul     Aug     Sep
#> 1   0.8101 -1.3260  1.9869  0.8885 -1.6659  0.5198  0.7025  0.3863 -0.6298
#> 2  -2.4712  1.5274 -0.0469 -0.0625  0.8081  0.9485  0.4706  0.5908  0.5179
#> 3  -0.0757 -0.3823 -1.6581 -0.5776  0.4045  1.3319 -0.1963  0.2404 -0.2559
#> 4  -0.1097  1.0181  0.6942 -0.2433 -0.3417  0.6157  0.5736 -0.7630 -0.7368
#> 5   0.7932  0.1457 -0.6877  0.0950 -1.2893  1.4566  1.9048 -0.4852 -1.1606
#> 6  -0.1866  0.5710 -0.1476 -0.9201  0.0333 -1.9337  1.3703  0.0352  2.0531
#> 7  -1.0751 -1.3823  0.1447 -0.6112 -0.0919 -0.0249 -0.7206 -0.7014 -0.2282
#> 8   0.2189  0.9631 -0.4644  0.2772  0.1889 -0.6754  1.1044  1.0393  0.5627
#> 9  -0.7086 -1.3359 -0.5941 -0.6975 -0.0596 -1.0373  0.0900 -0.7604  1.7379
#> 10  1.5807 -0.8136 -0.5954  1.3206 -0.8297 -0.3244 -0.5179 -0.4404 -0.3973
#> 11  1.2252 -0.3518 -0.2773 -0.2183  0.6571 -0.1696 -0.1395 -0.7908 -0.4231
#> 12 -1.6005  0.1996 -1.6680 -0.3776 -0.5965 -0.7037 -0.1446 -1.9699  1.1342
#> 13 -0.8194  2.0470 -0.0345 -1.2506 -1.5479  0.4182 -0.1177 -0.4832 -0.9137
#> 14 -0.6152 -1.0869  1.7682  0.4561 -0.3631  1.0923  0.7861 -0.6254  2.1636
#> 15  0.3683  2.1484  0.0493  0.5687 -0.2372 -0.6543 -0.1611  0.6594  2.0522
#> 16  0.9792  0.1316 -1.3933  0.3952 -0.1705  1.3197 -0.7920 -0.4824 -1.1722
#> 17  0.8102 -1.3799 -0.9606  1.4213 -1.1002  0.5555 -0.4926  0.4090  1.7837
#> 18  2.4897  0.6575  0.4905  0.0311  0.4505 -0.2738  0.4050 -0.6192 -1.3985
#> 19 -1.3374  0.6921 -0.1448 -1.9285 -0.0091 -0.6981 -0.4055 -1.2479  1.7662
#> 20  0.2637 -1.0307 -2.9265 -0.6792  0.5789 -0.1056 -0.0955  0.5630 -1.2192
#> 21  1.0447  0.1770 -0.9200  1.1283 -2.0282 -0.5449 -0.2468 -0.1755  1.0192
#> 22 -0.2824  0.6691  1.1193 -2.4113  0.5249  0.2918 -0.4241  0.9037 -0.4288
#> 23  0.2973 -0.4347 -1.1990 -0.3625 -1.0101 -0.3345  0.3668 -0.8029  2.6401
#> 24  0.6694 -1.2434  0.8780  0.3588 -0.5217  1.2886 -0.7416 -0.6187 -1.1077
#> 25  0.9410 -0.3878  0.3412  0.2047 -1.3965  1.6982  0.0794  0.6985  0.9260
#> 26 -1.0580  1.8886 -0.3881  1.4749  1.4485  0.0884  0.4758 -1.0738  0.1665
#> 27 -0.7186 -0.0749  0.9703 -0.4015 -0.2687 -1.3197  0.2422  1.2680  1.4907
#> 28  0.4851 -0.4650  0.0516 -1.0432  0.6129 -0.4670 -1.5660 -0.4017  1.4844
#>        Oct     Nov     Dec
#> 1   1.4507  0.1223 -1.0146
#> 2  -0.4653 -2.2564  0.2390
#> 3  -1.5126 -0.4791 -0.4884
#> 4   0.2365  0.7087 -0.1202
#> 5   0.4635 -0.2728  2.3351
#> 6   0.3224 -0.8688  0.4450
#> 7   1.1096 -1.6285  0.5548
#> 8   1.4096  0.4396 -1.5011
#> 9   0.6160  0.8054  0.2694
#> 10  0.3861 -0.2890 -0.6134
#> 11 -0.0511 -0.6782  0.4661
#> 12 -1.6959 -0.5079 -1.1680
#> 13 -1.5769 -0.9393  0.9387
#> 14 -0.4091 -1.4445 -0.0496
#> 15  0.9436  1.1884 -1.3290
#> 16 -0.2314  0.4965 -0.6974
#> 17  1.5985 -0.4072 -2.3688
#> 18  0.9746  1.4516 -0.4456
#> 19 -0.3503 -1.6692  0.1421
#> 20  1.0995 -0.0080  0.6970
#> 21  0.1122  0.3922 -0.0879
#> 22 -0.2763  0.8090  1.6823
#> 23  0.1345  1.1146 -2.3356
#> 24 -0.6168  1.0041  0.4383
#> 25 -0.1704 -0.3767  0.9004
#> 26  1.0609 -0.0761 -0.5234
#> 27  0.6881 -1.8687  0.4629
#> 28 -0.4508  0.7643      NA

Created on 2025-02-22 with reprex v2.1.1