You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hi!
First of all, great library, thanks for open-sourcing it!
In my project, I'd like to apply custom confidence intervals to the ARIMA predictions.
In the Readme file, you have stated that the confidence intervals can be changed in the ForecastUtil class, presumably through changing its confidence_constant_95pct constant. This, however, is not feasible if the project is used as a Maven dependency instead of using it as a local source dependency. It is, furthermore, not obvious to the layperson (or at least to me) how to calculate the constant for another confidence interval.
As far as I see it, the problem could be addressed in two ways ways:
Provide a public, static conversion method to get the constant for a custom percentile in ForecastUtil which can then be used with the setConfInterval method of the ForecastResult class. This would maybe be the simplest solution!
Extend the ArimaParams class with a confidence interval property, there could then maybe be another constructor with which custom percentiles can be set.
I'd be glad to contribute a solution to this if someone could give me some pointers on how to calculate the constant.
Have a great day!
The text was updated successfully, but these errors were encountered:
Hi!
First of all, great library, thanks for open-sourcing it!
In my project, I'd like to apply custom confidence intervals to the ARIMA predictions.
In the Readme file, you have stated that the confidence intervals can be changed in the
ForecastUtil
class, presumably through changing itsconfidence_constant_95pct
constant. This, however, is not feasible if the project is used as a Maven dependency instead of using it as a local source dependency. It is, furthermore, not obvious to the layperson (or at least to me) how to calculate the constant for another confidence interval.As far as I see it, the problem could be addressed in two ways ways:
ForecastUtil
which can then be used with thesetConfInterval
method of theForecastResult
class. This would maybe be the simplest solution!ArimaParams
class with a confidence interval property, there could then maybe be another constructor with which custom percentiles can be set.I'd be glad to contribute a solution to this if someone could give me some pointers on how to calculate the constant.
Have a great day!
The text was updated successfully, but these errors were encountered: