Future Support (2024)

[PavelRechkalov]

Greetings sir,

Are there any plans to continue developing / updating this library?

Kind regards and happy holidays

[hippke]

Yes, if there is a good case of what needs to be developed?

[PavelRechkalov]

Greetings sir,

Here are some ideas of trend filtering:

• Kalman Filter
• Particle Filter
• Robust STL: https://arxiv.org/abs/1812.01767
• Fast RobustSTL: https://www.researchgate.net/publication/343780200_Fast_RobustSTL_Efficient_and_Robust_Seasonal-Trend_Decomposition_for_Time_Series_with_Complex_Patterns
• Trend extraction based on VMD, EEMD, Wavelets

Moreover, it'd be great to enhance the Docs:

• Make clear what filters are "One-sided" or "Two-sided". So, potential users can understand what methods might have future information leakage problem (e.g, it's very important for finance applications)

Regards