compare_forecast_performance_per_group: Create plot to compare forecast performance per group

In ing-bank/tsforecast: Time Series Forecasting Pipeline

Description

compare_forecast_performance_per_group is a function to create a plot which compares the forecast performance of a single forecast model for every available group, based on a selected performance summary variable. For demo purposes, sensitive figures can be hidden from the audience.

Usage

compare_forecast_performance_per_group(
  accuracy_overview,
  fc_model = "",
  demo_mode = FALSE
)

Arguments

accuracy_overview	A tibble containing an overview of the overall forecast accuracy for each group in the main_forecasting_table, limited to data for a single forecast model.
fc_model	A character string specifying which forecast model to compare accross the available groups.
demo_mode	Boolean, which is to be set to TRUE if any potentially sensitive figures should be hidden from the audience for demo purposes, or set to FALSE if all figures can safely be displayed.

Value

A plotly object displaying performance per forecast model.

Examples

accuracy_overview <- tstools::initialize_ts_forecast_data(
      data = dummy_gasprice,
      date_col = "year_month",
      col_of_interest = "gasprice",
      group_cols = c("state", "oil_company")
   ) %>%
   create_main_forecasting_table(
      min_train_periods = 180
   ) %>%
   add_fc_models_to_main_forecasting_table(
      periods_ahead = 12,
      fc_methods = c("linear")
   ) %>% 
   get_forecast_accuracy_overview()
compare_forecast_performance_per_group(
   accuracy_overview = accuracy_overview,
   fc_model = "fc_linear_trend_seasonal"
)

ing-bank/tsforecast documentation built on Sept. 18, 2020, 9:40 a.m.