-
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
-
Thank you for your reply. here is an example of the bug i get :
" I used to forecast daily positive time series using tbats. But in some cases i get négatives or very long values like 1.25868e+14 whe…
-
I write this issue because tbats is behaving strangely and I think it's a bug.
File: https://dl.dropboxusercontent.com/u/1118905/bug.RData
Source:
```
library(forecast)
load("bug.RData")
y
-
TBATS gives an error with the simple sequence of zeros.
Repro:
{library(forecast);tbats( rep(0,10) )}
Result:
Error in checkForRemoteErrors(val) :
3 nodes produced errors; first error: initial va…
-
Hi, can it be that tbats somehow rvereses the time series? When I do plot(forecast(tbats,h=365)$mean),
the plot seems to be reveresed from the expected one...
I supply my time series in increasing ord…
-
Hi Rob,
I tried to measure accuracy of the "tbats" model using the code below and I got an error message. Does the accuracy function work for the tbats function? What would be a simple way to measure…