This paper has been thoroughly discussed, and I think even debunked, due to authors making a few mistakes and wrong assumptions. I hope I remember this correctly (feel free to correct me if I'm wrong), but they assumed that large wallets belonging to exchanges and online wallet services were wallets of indiviidual wealthy owners, as opposed to belonging to many different owners, and even possibly lumped user's personal wallets together, if the users used the same address to trade on these exchanges. They also didn't take into account that wealthy bitcoin owners break up their wealth among many smaller accounts, and they linked a bunch of addresses together with the assumption that if one address sends money to another, both belong to the same person, despite the transaction possibly being between two people.