-
This post and a project I'm working on for generating many forecasts in an automated manner has me thinking about selecting between ets(), auto.arima(), and tbats() models in an automated manner and p…
-
-
When a bats or tbats object, It would be nice to use the coefficient structure of these models unchanged in order to apply to a new series, maybe to compare actual outcome and prediction. In other wor…
-
Allow a specific model to be specified in both bats and tbats.
Also allow a fitted model to be reused, as in ets and Arima.
-
Reported at http://stackoverflow.com/q/32693896/144157
```
library(forecast)
weeklyu
-
To compute forecasts and fitted values without re-estimating the model.
-
Other forecast functions (e.g. `forecast.Arima()` or `forecast.ets()`) have checks on the `level` argument to prevent prediction intervals with negative levels or levels above 100. However, `forecast.…
-
Is it possible to have a function able to extract ets() components (level, slope, sesaonality) like tbats.components does with tbats() function?
-
Use `forecast::tbats` as explained here: http://robjhyndman.com/hyndsight/seasonal-periods/
For multi-seasonality, use `msts(x, seasonal.periods=c(1440,365)` or an `ets()` model
-
fit