[2019-01-02 Update]
Removed tsibble development and installed tsibble_0.8.5 from CRAN.
Session Info
R version 3.6.2 (2019-12-12)
Platform: x86_64-pc-linux-gnu (64-bit)
Running under: Ubuntu 19.10
tsibble_0.8.5 fable_0.1.1 fabletools_0.1.1
fit now shows <S3: lst_mdl> from the cell output.
At R console,
> fit
# A mable: 1 x 1
ets
<model>
1 <ETS(A,A,N)>
It looks likes an untrained model.
[Original]
On learning to fit time-series models with synthetic data, I used code based on intro to fable, and create a tsibble to make this example code:
library(fable)
library(tsibble)
fit <- tsibble(
date = as.Date("2017-01-01") + 0:9,
value = rnorm(10),
index = date
) %>%
model(ets = ETS(value ~ trend("A")))
fit
The above code shows error
1 error encountered for ets
[1] No common type for `..1` <interval> and `..2` <double>.
Searching "No common type" in fabletools suggests to me the message, and the error, comes from another package. A google search suggested it was tidyr, but it looks like this pull request changed it.
traceback() showed an unrelated error that originated from earlier in the session.
What can one do to further investigate this error?
What can I do to get the model to fit to the data?
