Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Using Benford’s Law to Detect Bitcoin Manipulation

statmodeling.stat.columbia.edu

Post reply on HN
Print view View original

61–70 of 96 posts

Re: Using Benford’s Law to Detect Bitcoin Manipulation

#61

Post by FabHK »

conformist wrote: post #16
dannyw wrote: post #7

What is the statistical likelihood of this appearing by chance? Note that 2014 to today isn't actually that long of a timeframe, as prices in a time-series are highly linked to the previous data point. Berkshire Hathaway has been trading for several decades; so that's 40 years of data; as compared to 7 years of data. yet the author, by using the same charts, seems to falsely imply that these are remotely comparable.…

Isn't Andrew Gelman's take more like "Look what they did, that's kind of interesting, what do you think?" rather than "look at this article, it is true"? See also the comment section.

Yes indeed. Most of the article is quoted from Smith, and Gelman (the prolific author of the worthwhile blog at columbia.edu) only really adds a few inconclusive sentences at the end. ("... There’s a lot going on, and people have lots of reasons for doing things.")

Re: Using Benford’s Law to Detect Bitcoin Manipulation

#62

Post by xiphias2 »

nootropicat wrote: post #28
xiphias2 wrote: post #13

Earlier quoted context omitted.

El Salvador's Bitcoin Beach project is a great example of people profiting from a debit based value network instead of a credit based financial system: even though both US and El Salvador had USD as the legal tender, it took $10 to remit $50 through the western union network. Lightning network (which is using Bitcoin as a settlement network) is both dramatically lowering the fees and provides instant debit transfer f…

LN isn't actually used - the way it actually works is: Strike holds dollars for the user. The user wants to send dollars to someone. Strike buys btc using dollars (from itself), sends btc via ln - to itself - and then sells btc for the same amount of dollars, again to itself, and credits the receiver. What's actually happening is that Strike is a normal payment provider like Paypal, Venmo, Revolut that fakes the btc…

You are totally right in each step, but there's 1 thing that makes it different from most exchanges: you can already transfer money in/out of it using LN, which makes it much more practical than using a Coinbase exchange account.

Also, while Coinbase Pro has a 0.1-0.3% fee for market order, if you want guaranteed execution, Coinbase and its competitors take 3-4%.

If you listened to talks from Jack Mallers, he was tracking lnd head instead of waiting for the full releases, and picking and testing pull requests by hand, fixing bugs that come up in practice (or giving feedback to lightning devs), that's one of Strike's advantages over other exchanges.

Also the ,,less regulation'' is just not true generally, people have to go through KYC to link with bank accounts, they are communicating with regulators: https://jimmymow.medium.com/announcing-strike-public-beta-32...

Re: Using Benford’s Law to Detect Bitcoin Manipulation

#63

Post by FabHK »

fluidcruft wrote: post #21
qPM9l3XJrF wrote: post #10

I don't buy it. My intuition here is that Benford's law is a thing because for pretty much any statistical distribution, large numbers are less likely than small numbers, and small numbers are more likely to start with 1. But I'd only expect this effect to show up when aggregating across many different statistical distributions (e.g. looking at all stocks in the S&P 500 at once) rather than looking at individual dist…

I don't really know anything about this but every time I read these analyses using Benford's Law I don't understand why anyone would expect Bitcoin's exchange rate to begin with the digit "1" ~30% of the time. Once you're not talking about human-guestimated numbers, it seems more like a question of scale factors. If bitcoin's value fluctuates between 30k and 70k it's never going to start with a "1". Or if its value f…

Benford's law tends to fit better when:

* the numbers span many orders of magnitude

* the numbers are produced by multiplication (where log(x) is normal-ish), not addition (where x is normal-ish)

Your examples span half an order of magnitude, so it definitely doesn't apply (as you say).

Re: Using Benford’s Law to Detect Bitcoin Manipulation

#64

Post by TomSwirly »

CrazyStat wrote: post #38
mumblemumble wrote: post #23

Earlier quoted context omitted.

The author at statsmodeling.stat.columbia.edu is the person who wrote several of the books on things like this. I don't think he can be finger-wagged away quite so easily as that. There's a huge difference between this and the election one, which is that this one is working with data that satisfies the key statistical assumptions needed to properly apply Benford's Law, and the QAnon elections rigged claims didn't. (O…

The time series spanning 3 orders of magnitude over a course if years is not sufficient condition for Benford's law to apply. As a trivial example, a stock whose price started at $1 and increased by $1 every day for 30 years would span 4 orders of magnitude over that time period, but would not follow Benford's law. A formal justification for applying Benford's law to a time series like that would depend on some kind…

> As a trivial example, a stock whose price started at $1 and increased by $1 every day for 30 years would span 4 orders of magnitude over that time period, but would not follow Benford's law.

You would expect if you deliberately created a security whose growth rate monotonically tends to zero over time that you would get anomalous results in any statistical point, but I don't quite see why you bring up this pathological example?

Re: Using Benford’s Law to Detect Bitcoin Manipulation

#66

Post by FabHK »

casi wrote: post #22
louwrentius wrote: post #2

Frankly, I can’t judge the merits of this article because I lack the knowledge required. But I think at this point we know that all cryptocurrencies are ‘greater fool’ “investments”. [1] They are in every way totally irrelevant and detrimental to society. Governments are unfortunately slow to crack down on the exchanges, although progress is being made. And that effort is essential to battle the cancer that is the ra…

I equally find it strange that people continue to come to hackernews- the Silicon Valley VC startup land - and can’t grasp that the infrastructure for programmable money might have some value. And write off p2p communication and coordination tools as zero-sum/ fraudulent games. And then wish for the government to ban other people’s jobs and hobby’s and communities because they don’t like it. Blows my mind that people…

You can easily (!) and efficiently (!) do "programmable money" without crypto[1]/blockchain/consensus (ie whatever Bitcoin and its descendants brought to the table), and guess what, people do.

> Blows my mind that people might spend their day coding, and night playing mmorpgs, and still not understand crypto.

Let's agree that you can lambast cryptocurrencies even despite understanding them, ok?

[1] with good old cryptography, of course, but without "crypto"

Re: Using Benford’s Law to Detect Bitcoin Manipulation

#67

Post by Etheryte »

Geee wrote: post #65
echopurity wrote: post #42

>Since bitcoins generate no income, their intrinsic value is zero Embarrassingly bad.

Yeah, it's like dollars or gold, their value is zero too. There is no such thing as 'intrinsic value'. Value is always subjective.

Intrinsic value is a specific term in finance [0] and it is by definition an estimation/approximation. Saying there's no such thing as intrinsic value because value is subjective is like saying there's no variables that are undefined because undefined variables have a defined value of undefined.

[0] https://www.investopedia.com/terms/i/intrinsicvalue.asp

Re: Using Benford’s Law to Detect Bitcoin Manipulation

#68

Post by FabHK »

dannyw wrote: post #7

What is the statistical likelihood of this appearing by chance? Note that 2014 to today isn't actually that long of a timeframe, as prices in a time-series are highly linked to the previous data point. Berkshire Hathaway has been trading for several decades; so that's 40 years of data; as compared to 7 years of data. yet the author, by using the same charts, seems to falsely imply that these are remotely comparable.…

Given that Benford's law tends to apply when the numbers span many orders of magnitudes, I would have expected more emphasis on that. Berkshire Hathaway (low 200, high 450,000) covers about 3.4 orders of magnitude (and much more time, ie opportunities to wander up and down), BTC (low 200, high 60,000) covers about about 2.5 orders of magnitude.

So, we should expect BTC to have a worse fit than Berkshire. How much worse would require some more sophisticated analysis than looking at the graphs, it seems.

Re: Using Benford’s Law to Detect Bitcoin Manipulation

#69

Post by Geee »

Etheryte wrote: post #67
Geee wrote: post #65

Earlier quoted context omitted.

Yeah, it's like dollars or gold, their value is zero too. There is no such thing as 'intrinsic value'. Value is always subjective.

Intrinsic value is a specific term in finance [0] and it is by definition an estimation/approximation. Saying there's no such thing as intrinsic value because value is subjective is like saying there's no variables that are undefined because undefined variables have a defined value of undefined. [0] https://www.investopedia.com/terms/i/intrinsicvalue.asp

It makes sense for something that produces cash flow, but saying that bitcoin doesn't have intrinsic value and therefore no value is just stupid.

Re: Using Benford’s Law to Detect Bitcoin Manipulation

#70

Post by mrfredward »

JumpCrisscross wrote: post #57
mrfredward wrote: post #47

Earlier quoted context omitted.

I guess we need to make a distinction between the blog post and the Gary Smith post it links to here. Gary smith (the person I think you're referring to having spent a career in this) says this: >The market manipulation, the irrational price gyrations, and the enthusiasm of so many investors for investing in bitcoin (and other cryptocurrencies) is ample evidence that market prices are not invariably equal to intrinsi…

> the world is full of people who feel very different paying $100.00 vs $99.99 Agree, though that effect is not constrained to Bitcoin. Retail orders, for instance, follow Benford's law. This is despite well-documented psychological biases towards e.g. certain digits, whole numbers, round numbers, et cetera [1]. Benford's law [2] derives from deeper mechanics. As you point out, however, a better control would have be…

Liquid retail stocks follow Benford's law because there is a notion of intrinsic value from the company and an army of quant traders trying to exploit any price inefficiency caused by the retail traders.

With cryptocurrency, the market is less mature and the intrinsic value largely comes from people believing in its value. So really, it would be surprising if we didn't see some Benford's law anomalies associated with people picking numbers.

Anyway, thanks for the discussion; the links above have given me stuff to chew on and calmed the red mist after I got so many drive-by downvotes.

Post reply on HN