Mitchell O'Hara-Wild
Mitchell O'Hara-Wild
```r coverage
When a model is linked with others in a list (possibly with some reconciliation/aggregation constraints) it should be possible to bootstrap the series using the same time point across all...
Related: https://github.com/earowang/hts/pull/51
Currently it drops the fable class, even if the renamed columns aren't special in anyway (key, index, distribution). It should be possible to rename all columns without issue, maintaining the...
Ref: https://stackoverflow.com/questions/70279495/select-a-single-forecast-plot-using-fableautoplot
Due to the 'dummy forecasts' used to represent the upper levels of the hierarchy, these dummies may not correctly represent the expected structure of the aggregated forecasts.
The current system of reusing a single R6 class for each slice of the data (key) has a few issues: 1. Cannot easily choose models by row (https://github.com/tidyverts/fable/issues/54) 2. More...
https://stackoverflow.com/questions/68747706/error-cant-subset-columns-that-dont-exist-selecting-from-a-fable-in-r/68751090?noredirect=1#comment121524319_68751090
Currently each model created in `estimate()` contains no key variables, which limits the ability to use global models. Adding the key dimension to modelling requires substantial work in generalising methods...
Currently the order of plotting intervals is not customisable. Instead it should respect the factor order if available.