fabletools

fabletools copied to clipboard

When using custom transformation functions (like `my_scaled_logit()` below), the modeling functions like `ETS()` or `ARIMA()` seem to search only the global environment for their definition. When run inside a different...

hplieninger

Using a multivariate response with univariate models could estimate each response variable independently using the same model structure. https://stackoverflow.com/questions/67051079/r-forecasting-fable-package-multivariate-decompostion-model

mitchelloharawild

forecast with reconciliation does not work in time series with only one non-zero observation

1

I am trying to run an experiment where I have time series that contains only one non-zero observation. Th experiment runs for the following example where I have two non-zero...

bahmanrostamitabar

Provide xreg specials helpers

1

xreg specials typically error or call `model.frame()` for an interface consistent with `lm()`. Rather than requiring model developers to rewrite this common pattern for each model, providing helpers in fabletools...

mitchelloharawild

Convert null_model to be simply NULL

3

There are many cases in which NULL can be generated from other tidyverse tools. Handling NULL models should be handled with the simpler `NULL` value.

mitchelloharawild

Add support for distributions of fitted values

1

mitchelloharawild

.model name clash when forecasting from components()

3

Hi, When forecasting off the back of a `components()` call, you run into an issue because there is already a '.model' column. I ran into this when running through Chapter...

gregfoletta

mitchelloharawild

Add model bagging

2

`forecast::baggedModel()` It should accept a model definition for decomposition, and then use a time series bootstrap to produce bootstrapped residuals for the decomposition, which are then recombined to give bootstrapped...

mitchelloharawild

Graceful failure when forecasting a mable

2

Currently running forecast on a mable fails if any of the models errors out when being forecasted. Is there a way to either drop it completely or return some "NULL...

Fuco1