Mitchell O'Hara-Wild
Mitchell O'Hara-Wild
That sounds promising. There will need to be more work than I had hoped to get `mutate()`, `filter()`, etc. to behave as expected - so a new class with dplyr...
While I think this functionality is useful, I want to wait to see more developments with slurrr (#143) before working on this further. The original motivating issue with the window...
If we were to have a family of `roll_by()` functions I'd expect them to work with `mutate()`, `filter()`, etc. To achieve this a subclass would be required unfortunately (upon confirming...
This is because `stats::stl()` only works for seasonal series, and we fall back to `stats::supsmu()` for non-seasonal cases. Removing this (bad) functionality would be too much of a breaking change,...
`...` isn't passed through to `stats::supsmu()`, it is passed to `stats::stl()` instead.
Closing as the misunderstanding seems resolved, and is now a duplicate of https://github.com/tidyverts/feasts/issues/62
Do you have an example dataset to better describe what you would like to make? If you save the gg_tsdisplay output, you are able to modify each individual plot. For...
Perhaps a suggested package as the functionality is available without it.
Wait for https://github.com/mitchelloharawild/moment to provide origin-less time structures
Minor MRE fix: to create a tsibble, the `index` argument should match a column name, not a value: ``` r library (tsibble) library (units) #> udunits system database from /usr/share/xml/udunits...