Description Usage Arguments Value Author(s) Examples
Regression based models use a linear trend to account for the change in level over time. In practical terms, it is measured as a vector of equidistant integers. Often, the trend component can significantly impact the forecast in the long run. One way to solve this issue is to apply a "discount" on the trend vector, henceforth, reducing the marginal effect on the predictions.
1 | get_trend_decay(y_var_length, trend_decay, horizon = NULL, lag = NULL)
|
y_var_length |
integer: Length of the time series |
trend_decay |
numeric: How rapidly the trend reach the stability. |
horizon |
integer: How far in time to produce trend discounts. |
lag |
integer: Lag to be used for cross-validation purposes. |
Numerical vector.
Obryan Poyser
1 2 3 4 | ## Not run:
get_trend_discounts()
## End(Not run)
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.