Skip to contents

Calculate forecast errors and percentage errors

Source: R/make_errors.R
make_errors.Rd

Calculate forecast errors and percentage forecast errors for point forecasts.

Usage

make_errors(future_frame, main_frame, context)

Arguments

future_frame

A tibble containing the forecasts. It must contain the columns specified by context, as well as model, split, horizon, and point.

main_frame

A tibble containing the observed values. It must contain the series identifier, time index, and value column specified by context.

context

A named list with the identifiers for series_id, value_id, and index_id.

Value

A tibble containing forecast errors and percentage forecast errors.

Details

make_errors() compares point forecasts in future_frame with the observed values in main_frame. The two data sets are joined by the series identifier and time index specified in context.

The forecast error is calculated as error = actual - point. The percentage forecast error is calculated as pct_error = (actual - point / point) * 100.

Positive errors indicate that the forecast is below the observed value. Negative errors indicate that the forecast is above the observed value.

The returned data contains:

  • series_id: Unique identifier for the time series as specified in context.

  • model: Forecasting model name.

  • split: Train-test split identifier.

  • horizon: Forecast horizon.

  • error: Forecast error.

  • pct_error: Percentage forecast error.

See also

Other accuracy functions: mae_vec(), make_accuracy(), mape_vec(), me_vec(), mpe_vec(), mse_vec(), rmse_vec(), smape_vec()

Examples

library(dplyr)
library(tsibble)
library(fabletools)
library(fable)

context <- list(
  series_id = "series",
  value_id = "value",
  index_id = "index"
)

main_frame <- M4_monthly_data |>
  filter(series %in% c("M23100", "M14395"))

split_frame <- make_split(
  main_frame = main_frame,
  context = context,
  type = "first",
  value = 120,
  n_ahead = 18,
  n_skip = 17,
  n_lag = 0,
  mode = "stretch",
  exceed = FALSE
)

train_frame <- slice_train(
  main_frame = main_frame,
  split_frame = split_frame,
  context = context
) |>
  as_tsibble(
    index = index,
    key = c(series, split)
  )

model_frame <- train_frame |>
  model(
    "SNAIVE" = SNAIVE(value ~ lag("year"))
  )

fable_frame <- model_frame |>
  forecast(h = 18)

future_frame <- make_future(
  fable = fable_frame,
  context = context
)

error_frame <- make_errors(
  future_frame = future_frame,
  main_frame = main_frame,
  context = context
)

error_frame
#> # A tibble: 90 × 6
#>    series model  split horizon    error pct_error
#>    <chr>  <chr>  <int>   <int>    <dbl>     <dbl>
#>  1 M14395 SNAIVE     1       1  -279.    -15.0   
#>  2 M14395 SNAIVE     1       2  -431.    -23.5   
#>  3 M14395 SNAIVE     1       3 -1624.    -99.6   
#>  4 M14395 SNAIVE     1       4  -573.    -34.5   
#>  5 M14395 SNAIVE     1       5    -1.10   -0.0708
#>  6 M14395 SNAIVE     1       6   103.     10.1   
#>  7 M14395 SNAIVE     1       7   -15.6    -2.18  
#>  8 M14395 SNAIVE     1       8  -409.    -42.8   
#>  9 M14395 SNAIVE     1       9  -257     -21.0   
#> 10 M14395 SNAIVE     1      10  -291.    -34.9   
#> # ℹ 80 more rows