Skip to contents

Calculate Accuracy Metrics

Source: R/modeltime-accuracy.R
modeltime_accuracy.Rd

This is a wrapper for yardstick that simplifies time series regression accuracy metric calculations from a fitted workflow (trained workflow) or model_fit (trained parsnip model).

Usage

modeltime_accuracy(
  object,
  new_data = NULL,
  metric_set = default_forecast_accuracy_metric_set(),
  acc_by_id = FALSE,
  quiet = TRUE,
  ...
)

Arguments

object

A Modeltime Table

new_data

A tibble to predict and calculate residuals on. If provided, overrides any calibration data.

metric_set

A yardstick::metric_set() that is used to summarize one or more forecast accuracy (regression) metrics.

acc_by_id

Should a global or local model accuracy be produced? (Default: FALSE)

  • When FALSE, a global model accuracy is provided.

  • If TRUE, a local accuracy is provided group-wise for each time series ID. To enable local accuracy, an id must be provided during modeltime_calibrate().

quiet

Hide errors (TRUE, the default), or display them as they occur?

...

If new_data is provided, these parameters are passed to modeltime_calibrate()

Value

A tibble with accuracy estimates.

Details

The following accuracy metrics are included by default via default_forecast_accuracy_metric_set():

  • MAE - Mean absolute error, mae()

  • MAPE - Mean absolute percentage error, mape()

  • MASE - Mean absolute scaled error, mase()

  • SMAPE - Symmetric mean absolute percentage error, smape()

  • RMSE - Root mean squared error, rmse()

  • RSQ - R-squared, rsq()

Examples

library(tidymodels)
library(dplyr)
library(lubridate)
library(timetk)


# Data
m750 <- m4_monthly %>% filter(id == "M750")

# Split Data 80/20
splits <- initial_time_split(m750, prop = 0.8)

# --- MODELS ---

# Model 1: prophet ----
model_fit_prophet <- prophet_reg() %>%
    set_engine(engine = "prophet") %>%
    fit(value ~ date, data = training(splits))
#> Disabling weekly seasonality. Run prophet with weekly.seasonality=TRUE to override this.
#> Disabling daily seasonality. Run prophet with daily.seasonality=TRUE to override this.


# ---- MODELTIME TABLE ----

models_tbl <- modeltime_table(
    model_fit_prophet
)

# ---- ACCURACY ----

models_tbl %>%
    modeltime_calibrate(new_data = testing(splits)) %>%
    modeltime_accuracy(
        metric_set = metric_set(mae, rmse, rsq)
    )
#> # A tibble: 1 × 6
#>   .model_id .model_desc .type   mae  rmse   rsq
#>       <int> <chr>       <chr> <dbl> <dbl> <dbl>
#> 1         1 PROPHET     Test   263.  356. 0.813