I am unable to let the ACCURACY function point only to the Validation_set. The code perfectly works only in case the first argument of Accuracy is the whole dataset(“data_tsibble”) but in case I indicate only the validation_set i receive following error message that in English means “the argument must be cohercible to a non-negative integer”:

Messaggi di avvertimento:
1: 2 errors (1 unique) encountered
[2] l'argomento dev'essere coercibile in un intero non-negativo
 
2: 2 errors (1 unique) encountered
[2] l'argomento dev'essere coercibile in un intero non-negativo
 
3: 2 errors (1 unique) encountered
[2] l'argomento dev'essere coercibile in un intero non-negativo
 
4: 2 errors (1 unique) encountered
[2] l'argomento dev'essere coercibile in un intero non-negativo
 
5: 2 errors (1 unique) encountered
[2] l'argomento dev'essere coercibile in un intero non-negativo

Here follows the full code

# Clear all variables from workspace
rm(list = ls())

# Load required packages
library(fpp3)
library(readxl)
library(tidyverse)
library(fable)
library(fabletools)
library(tsibble)
library(prophet)
library(forecast)
library(urca)
library(distributional)
library(lubridate)
library(writexl)
library(ggplot2)

# Read the dataset
data <- read_excel("D:/TEST.xlsx")

# Convert the data into a tsibble (time series tibble)
data_tsibble <- data |>
  mutate(Mesi = yearmonth(Mesi)) |>
  as_tsibble(index = Mesi)

# Visualize plot of the original series
data_tsibble |>
  autoplot(Intermediato) +
  xlab("Mesi") + ylab("Intermediato")

# Create training set and validation set
train_set <- data_tsibble |> filter_index(~ "12/01/2023")
validation_set <- data_tsibble |> filter_index("01/01/2024" ~ .)


fit_models <- train_set |>
  model(Ets = ETS(Intermediato),
        Arima = ARIMA(Intermediato, stepwise=FALSE, approximation=FALSE),
        tslm1 = TSLM(Intermediato ~ trend() + season()),
        tslm2 = TSLM(Intermediato ~ trend() + I(trend()^2) + season()),
        tslm3 = TSLM(Intermediato ~ trend() + I(trend()^2) + season()))
glance(fit_models) |> arrange(AICc) |> select(.model:BIC)


fc_models <- fit_models |>
  forecast(h=nrow(validation_set))

# If the first argument of the function ACCURACY is the whole dataset (data_tsibble), everything works out fine

fc_models |>
  accuracy(
    data = validation_set,
    measures = list(
      crps = CRPS,
      rmse = RMSE,
      mase = MASE,
      ss_crps = skill_score(CRPS),
      ss_rmse = skill_score(RMSE)
    ),
    n_quantiles = 100
  ) |>
  group_by(.model) %>%
  summarise(
    ss_crps = mean(ss_crps) * 100,
    ss_rmse = mean(ss_rmse) * 100,
    mase = mean(mase)
  )


# If the first argument of the function ACCURACY is the whole dataset (data_tsibble), everything works out fine
fc_models |>
  accuracy(validation_set,
           list(rmse=RMSE,
                mae=MAE,
                mape=MAPE,
                mase=MASE,
                rmsse=RMSSE,
                acf1=ACF1,
                crps=CRPS,
                skill_crps=skill_score(CRPS),
                skill_rmse=skill_score(RMSE)))

How can i solve this problem?

Thanks a lot

Mark