Hello. I'm looking for a fast way to group by similarity many (5-10k) long (2-10k points) time series. Using PAM on distance matrix obtained via as.dist(1-abs(cor(data))) produces usable results but it's rather slow and doesn't catch slightly shifted time series. DTW implementation from package 'dtw' is orders of magnitude slower even with global window constraints which renders it unusable for me.
Is there any fast and robust to noise and slight misalignment time series comparison implementation for R? There are fast algorithms that are able to deal with large time series http://www.cs.ucr.edu/~eamonn/iSAX/iSAX.html http://www.cs.ucr.edu/~eamonn/UCRsuite.html Is there any SAX, UCR DTW, UCR ED implementations for R? Regards, Anton Lebedevich. ______________________________________________ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
