Timeline for Check if my time series is forecastable using Shannon entropy
Current License: CC BY-SA 4.0
5 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Apr 29 at 0:44 | comment | added | Marco | ok, thank you very much, I will try to do it. | |
| Apr 28 at 23:45 | comment | added | Cryo | @Marco, I don't know whether it makes sense. I don't know the exact implementation of your algorithm, and even if I did I don't use entropy that often, and even if I did, I don't know the entropy in your time-series (before the spectral bit etc). This is precisely what I meant by my comment in the OP. You have got the number 'Now what?'. As I explained above, I think you need to simulate/synthesize some time-series with properties that you control and put them into your entropy computing algorithm. This will give you a reference point, i.e. what normal entropy for you looks like. | |
| Apr 28 at 23:01 | comment | added | Marco | I'm sorry for the delay in answering to you. I appreciate your help but I'm having trouble to understand this. I recently started using spectral entropy to analyze time series, following the information I provided. I'm having difficulty for interpreting the results, the entropy of the last 25% of a series is 0.19, and the entropy of the entire series is 0.23, does this make sense? Isn't the more data a neural network has better to forecast, for example? | |
| Mar 28 at 7:51 | comment | added | Cryo | Quick link on the inverse of a Lag operator, in case it helps ealdrich.github.io/Teaching/Econ211C/LectureNotes/Unit1-ARMA/… | |
| Mar 28 at 7:50 | history | answered | Cryo | CC BY-SA 4.0 |