I have a number of non-periodic signals , that have been detrended and denoised. Each signal is made up of a variable number of cycles non necessarily of the same length. We can now fit a regression model to each cycle and then use the returned coefficients, together with some other data, to form clusters.
We would like to keep track of which signal each cycle came from in the clusters layout. It must be a trivial problem. However I cannot pass such identifier to the routines that calculate the similarity matrix as it would be used in the distance calculation process. So I ran out of ideas about how to keep the clustering objects (cycles) distinguishable all along the clustering process (calculation of similarity matrix and later clustering). Is that doable at all ? Thank you so much, -- Maura E.M ______________________________________________ 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.
