-
- don't be more precise than necessary (e.g. when methods use unique events times, data is given very precisely, but time scale of interest is very large, e.g. `hdfail` example)
-
@fkiraly update: `sktime` has a class `AutoETS` which interfaces `statsmodels` `ETSModel` but adds some auto-fitting functionality on top.
`statsmodels` `ETSModel` should also be exposed 1:1 as an …
-
https://github.com/josef-pkt/statsmodels/tree/REF_covtype_fit
misc notes:
- wls_prediction_std needs to be added to unit tests for robust cov.
after the first commit, I have a scale that is much to…
-
It would be great if support can be added to compute simultaneous confidence intervals, similar to routines available in emmeans, rms, etc.
-
I am trying to implement out of sample prediction using predict function of Pyflux package. But to my surprise instead of giving out of sample prediction, it returns in sample predicted values.Sample …
ghost updated
4 years ago
-
Our prediction intervals sometimes have a trend, suggesting that the residuals for that model have a non-zero null. This points to a sub-optimal fit - we should check for it and fix it.
-
I made an adaption of the ballots notebook for a similar binary problem (figuring out what the true percentage of happy customers are for a business.) In my case I find that the difference in width of…
-
Summarizes some optional features discussed for `FallbackForecaster` (see first version in https://github.com/sktime/sktime/pull/5779) for future work:
* currently, does not support probabilistic f…
-
As a universal approach for all methods in CLVTools, implement a boostrapping procedure to be able to generate confidence intervalls for predictions (CET/DERT/CLV)
-
Hi Team,
Is it possible for .predict function to generate strictly positive predictions and confidence intervals? If so can you please guide on the change the pmd auto arima and/or predict for this t…