Time series data

Question

I would like two compare two/more time series data (data are acquired slightly different times. acquired every other seconds for a month). And possibly find anomaly if there is any strange point in one of the data. How do i do that in python?

Answer 1

There are at least a few approaches to anomaly detection, of which I can recommend the machine learning-based ones. In particular, kNN should do the trick as it will show the outliers pretty quickly (the average or median of the distances to k nearest neighbors will be significantly higher for distant/remote/anomalous datapoints).

Answer 2

I managed to compare the time shift between the two data using numpy.fft.