search

laszukdawid / PyEMD

Python implementation of Empirical Mode Decompoisition (EMD) method
https://pyemd.readthedocs.io/
Apache License 2.0
857 stars 222 forks source link

QUESTION - Does pyEMD, CEEMDAN try to account for end effects? #135

Open ctanny opened 1 year ago

ctanny commented 1 year ago

I am curious to know if there is anything in the package, specifically in the CEEMDAN class that addresses the end effects issues with EMD? Thanks.

laszukdawid commented 1 year ago

Can you please elaborate what do you mean by "end effects issue with EMD?" EMD supports mirroring for time series beginning/end and CEEMDAN uses these algorithms when computing EMD. Is there anything else needed?

ctanny commented 1 year ago

I'm not sure. I have been experimenting with regression and LSTM models to help reduce the end effects. But if mirroring reduces these issues and is incorporated into the CEEMDAN algorithm, then thanks! That's a huge potential problem that is dealt with. can you point me to where mirroring is discussed in the docs? Thank you for sharing this great work.