Improved cvts - Githubissues

ellisp / forecastHybrid

Convenient functions for ensemble forecasts in R combining approaches from the {forecast} package

GNU General Public License v3.0

79 stars 23 forks source link

Improved cvts #68

Closed ganesh-krishnan closed 7 years ago

ganesh-krishnan commented 7 years ago

Saved cross validation params to the cvts object
Allowed additional arguments to be passed to fitting function via ...
Added a helper function tsCombine to combine time series with the same frequency
Added function to extract forecasts
- Can specify forecast horizon
- Will always return an object with the same time period as the original series for easy comparison. Values not present (because of chose horizon or lack of forecasts) will be na
Rewrote tests

ganesh-krishnan commented 7 years ago

Next steps:

[ ] Add support for xreg
[ ] Add plotting methods

ganesh-krishnan commented 7 years ago

Not really sure why the tests for accuracy are failing. They are passing on my local machine. I have not modified any code in that part of the package. Any ideas @dashaub

dashaub commented 7 years ago

@ganesh-krishnan The accuracy test failing was a bogus test I put in here to try to fix the Travis reporting (which still isn't working), so it isn't related to your PR. I left this open Friday and forgot to revert it, but really I shouldn't have put it in the master branch. As best as I can see there might be some documentation arguments that need to be added but the rest looks fine.