No applicable method for "accuracy" applied to an object of class "forecast" while knitting R

[sasun8820]

Hi! I am having difficulty knitting R with the following codes. The function accuracy () works properly to give its outputs in an rmd file, but when I attempted to knit it, I received the error depicted in the image below.

I thought accuracy is a function that takes both forecast and ts objects.. I have no clue what is going on here.
Any help will be much appreciated!

w_data = writing 

library(e1071)
library(caret)
library(forecast)
library(dplyr)

w_training = window(w_data, start=1968, end=c(1975,12))
w_test = window(w_data, start = 1976)

naive_test = stlf(w_training,t.window=13,lambda = BoxCox.lambda(w_training),s.window="periodic",robust=TRUE,method="naive",h=length(w_test))
accuracy(naive_test, w_test)

rwdrift_test = stlf(w_training,t.window=13,lambda = BoxCox.lambda(w_training),s.window="periodic",robust=TRUE,method="rwdrift", h=length(w_test))
accuracy(rwdrift_test, w_test)

(I got the output from above codes as below but keep having an error msg while knitting it)

[robjhyndman]

It's hard to tell what the error is without a reprex (FAQ: What's a reproducible example (`reprex`) and how do I create one?). Here is an example that works with your code.

library(forecast)
#> Registered S3 method overwritten by 'quantmod':
#>   method            from
#>   as.zoo.data.frame zoo

w_data <- ts(abs(rnorm(200)), start = 1968, frequency = 12)
w_training <- window(w_data, start = 1968, end = c(1975, 12))
w_test <- window(w_data, start = 1976)

naive_test <- stlf(w_training, t.window = 13, 
  lambda = BoxCox.lambda(w_training), s.window = "periodic", 
  robust = TRUE, method = "naive", h = length(w_test))
accuracy(naive_test, w_test)
#>                       ME      RMSE       MAE       MPE      MAPE      MASE
#> Training set  0.00394147 0.7864543 0.6299847 -4068.137 4128.5315 0.9773049
#> Test set     -0.17644808 0.6118244 0.5208971  -306.057  327.5739 0.8080756
#>                    ACF1 Theil's U
#> Training set -0.4930051        NA
#> Test set     -0.1100589  0.490973

rwdrift_test <- stlf(w_training, t.window = 13, 
  lambda = BoxCox.lambda(w_training), s.window = "periodic", 
  robust = TRUE, method = "rwdrift", h = length(w_test))
accuracy(rwdrift_test, w_test)
#>                        ME      RMSE       MAE        MPE      MAPE      MASE
#> Training set -0.001135496 0.7866437 0.6304382 -4105.1823 4164.9380 0.9780084
#> Test set     -0.454331138 0.7537172 0.6379784  -425.3502  436.0732 0.9897057
#>                     ACF1 Theil's U
#> Training set -0.49300666        NA
#> Test set     -0.05154857 0.5016373

^{Created on 2023-01-31 with reprex v2.0.2}

[sasun8820]

Thank you for your work! The data I pulled off the training & test set was writing series from the package fpp2? Unfortunately, even after using your codes, I still fail to knit it with the same error message.

[robjhyndman]

Please create a reprex: FAQ: What's a reproducible example (`reprex`) and how do I create one?

Otherwise there is no way to know what the issue is.

[sasun8820]

I followed the guideline. I am not sure if this is a way to be done. Thanks in advance

library(fpp2)
#> Registered S3 method overwritten by 'quantmod':
#>   method            from
#>   as.zoo.data.frame zoo
#> ── Attaching packages ────────────────────────────────────────────── fpp2 2.4 ──
#> ✔ ggplot2   3.4.0     ✔ fma       2.4  
#> ✔ forecast  8.20      ✔ expsmooth 2.3
#> 
w_data = writing 
head(w_data,3)
#>          Jan     Feb     Mar
#> 1968 562.674 599.000 668.516

tsdisplay(w_data)

autoplot(w_data)

ggsubseriesplot(w_data, main = NULL) 

^{Created on 2023-01-31 with reprex v2.0.2}

# Split into a 80-20 on training & test data 
w_training = window(w_data, start=1968, end=c(1975,12))
#> Error in window(w_data, start = 1968, end = c(1975, 12)): object 'w_data' not found
# w_training = head(w_data, 96)
w_test = window(w_data, start = 1976)
#> Error in window(w_data, start = 1976): object 'w_data' not found
# w_test = tail(w_data, 24)

# library(e1071)
# library(caret)
# library(dplyr)
library(forecast)
#> Registered S3 method overwritten by 'quantmod':
#>   method            from
#>   as.zoo.data.frame zoo
naive_test = stlf(w_training,t.window=13,lambda = BoxCox.lambda(w_training),s.window="periodic",robust=TRUE,method="naive",h=length(w_test))
#> Error in NCOL(x): object 'w_training' not found
accuracy(naive_test, w_test)
#> Error in accuracy(naive_test, w_test): object 'naive_test' not found

rwdrift_test = stlf(w_training,t.window=13,lambda = BoxCox.lambda(w_training),s.window="periodic",robust=TRUE,method="rwdrift", h=length(w_test))
#> Error in NCOL(x): object 'w_training' not found
accuracy(rwdrift_test, w_test)
#> Error in accuracy(rwdrift_test, w_test): object 'rwdrift_test' not found

^{Created on 2023-01-31 with reprex v2.0.2}

[robjhyndman]

This is two separate reprexes. As you can see from the errors, the second one can't access the data. If you put them in the same reprex, it should work.