Hacker News Reader

Comment by mr_spothawk | Hacker News Reader

It looks like the Gompertz function is asymptotic... i wonder about how they fit that model to the price spikes.

Also, stupid question: what's the expected price then, given that the infection nears that asymptote by say, 90%?

The study says that the price almost perfectly fits the curve, but they exclude the price spikes in that fit and label those periods as "manipulated price".

This strikes me as a rather strange approach to fitting data. But I'm not a data scientist, and they do give some references to other studies about bitcoin price manipulation to support this.

Isn't price manipulation the purpose of Bitcoin? The market is so small that you can actually corner it with very little effort, and since it's Bitcoin there's no one to enforce the rules.

No. According to the original author, the purpose of bitcoin is to provide an electronic payment system that does not require trusted third parties. See the introduction section of https://bitcoin.org/bitcoin.pdf

As of today the market cap of Bitcoin is $70 billion and the daily volume is $10 billion, so it takes a little effort. And regulated exchanges in the U.S. and other developed countries actually do have to follow some rules.

I didn't read the the study to too much detail, but based off the charts they seem to suggest many (not all of course) deviations from the expected value is due to manipulation.

To me this is in-line with the reports of market manipulation [0] and the inherit lack of regulation.

[0] https://hackernoon.com/meet-spoofy-how-a-single-entity-domin...

Reply on news.ycombinator.com