Hi
I have a large time series with two variables ( In reality more dependent variable, but now focusing on only two) in which variable2 is dependent on variable1. Algorithm need to identify whenever variable2 differ from its ideal path in dependent to variable1. Would like to know any efficient machine learning method suited for this application ( Which can be adapted in future for multiple variable ) in Matalb .
Simple plotting and visually inspecting is laborious because of the large data set.
I suggest you to the following links:
https://www.youtube.com/watch?v=My9XLzmxoRU
https://towardsdatascience.com/anomaly-detection-with-time-series-forecasting-c34c6d04b24a
https://anomaly.io/detect-anomalies-in-correlated-time-series/
You can apply RNN and LSTM, these deep learning tools are available in Matlab 2018 (or onward) .
- Badges
- Science topic
- Similar topics
- Statistics
- Time Series