Statistically Robust Distance Measure/Metric for comparing more than two network data series

I have about 30 lists of unequal length (some of which are triplicates of the data), corresponding to metrics relating to nodes of different graphs. I want to compare their similarity using a distance metric, but was unsure which method I can use given the data lists are of unequal length. I was exploring using dynamic time warping, but was wondering if there is any other more basic method.

For example, I was a considering creating histograms with same bin edges and number of bins for each list and using a distance metric on the frequency, but I am not sure how to go about this using python, or if there's a function/package that does this already. Is this even a "good" way of doing it?

I'm also interested in finding a way to measure the statistical significance of the distance measures between the different graphs.

This is a lot in one question, I am new to this and appreciate any help. Thank you in advance!

Just to clarify the question: Do the lists describe different graphs, or do you need the similarities to learn which lists refer to the same graph? Is data tri- or duplicated within single lists or are lists duplicating other lists? Would you consider removing redundant metrics?

Sorry, this was rather a comment than an answer, but I am not yet entitled to comment here, so I am starting the discussion like this.

• No need to apologize, thank you for asking! For each graph type, there are three "triplicate" type lists (also of different length) and I have such triplicates for many different graphs that I want to measure the the difference between. In case it is useful, the lists for each graph correspond to topological metrics for the nodes of the graphs (e.g. betweenness centrality). The triplicates are for (hopefully) obtaining a robust statistical measure of the difference between lists corresponding to different graph types, not the same graph. Mar 10, 2018 at 21:00