has_large_seasonal_component: Compare residuals to judge whether series has a sizeable...

In timradtke/heuristika: Robust Probabilistic Forecasts to Tinker With

Compare residuals to judge whether series has a sizeable seasonal component

Description

Compare residuals to judge whether series has a sizeable seasonal component

Usage

has_large_seasonal_component(residuals, residuals_after_seasonality, threshold)

Arguments

residuals	Residuals from prediction without seasonal component
residuals_after_seasonality	Residuals from prediction with seasonal component
threshold	A value between 0 and 1 describing the share of variation removed from the residuals due to inclusion of the seasonal component

Value

Logical scalar

Examples

residuals_after_seasonality <- rnorm(n = 50)
residuals <- residuals_after_seasonality + 5 * cospi((1:50) / 6)

plot(residuals, pch = 19)
lines(residuals, lty = 1)
points(residuals_after_seasonality)
lines(residuals_after_seasonality, lty = 2)

tulip:::has_large_seasonal_component(
  residuals = residuals,
  residuals_after_seasonality = residuals_after_seasonality,
  threshold = 0.5
)

timradtke/heuristika documentation built on April 24, 2023, 1:55 a.m.