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

Nassim Taleb: We should retire the notion of standard deviation

edge.org

Post reply on HN
Print view View original

161–170 of 254 posts

Re: Nassim Taleb: We should retire the notion of standard deviation

#161

Post by tn13 »

There is nothing wrong with STD or MAD. The real problem is a lot of people apply them without realizing the nature of their data and what kind of analysis they want to do.

In this case what matters in the end is the kind of impact deviation from mean has on the real world variable you have. I agree that in most Gaussian experiments MAD might be more useful than STD.

STD is more useful when the real world impact of the deviation increases exponentially with the magnitude of deviation and hence it is a good idea of magnify the (x-n) by squaring it. In many cases the impact is linear where MAD clearly works better. For example in cricket where n runs are n times better than 1 run. But in case of shooting. Hitting 9 targets out of 10 might be 100 times better than 1 out of 10 so there MAD will be misleading.

Re: Nassim Taleb: We should retire the notion of standard deviation

#162

Post by Helianthus »

jerf wrote: post #144
Helianthus wrote: post #141

Earlier quoted context omitted.

>Reality always provides some bound But... it doesn't. You ever hear about the hypothetical possibility of your atoms lining up and falling through the floor? It's hypothetical in the sense that it's really ridiculously unlikely, but there is no bound preventing it . Now the central point about different probability curves stands, but that's not what Taleb was talking about--he seems to think that it's the tool's fau…

"But... it doesn't. You ever hear about the hypothetical possibility of your atoms lining up and falling through the floor?" A bad example; that's a very, very large sample space, such that deviations from mathematical perfection are irrelevant. They do exist, if you're precise enough (for instance, the universe is not modeled by perfectly continuous space), but I'm not inclined to argue them, because it's too easy t…

>The fact that this isn't precisely what Homunculiheaded said is also why I posted

Ah. I misread the following...

>>This is actually what I expected to read:

as agreement ("This is actually what I expected to read."). My mistake.

>Unfortunately, rather more science takes place in the second domain than the first.

As I said to Homunculiheaded, this is because of the relative utility of the models, which we understand--and even those that do not understand it do not make the tool's use invalid.

What are we bemoaning, here, but actual misunderstanding itself?

And really, what's the point of that?

Re: Nassim Taleb: We should retire the notion of standard deviation

#163

Post by corresation »

kaa2102 wrote: post #119
corresation wrote: post #97

Earlier quoted context omitted.

There is no snark in my post: I am outright calling Taleb a false prophet who is guilty of exactly the things he criticizes in Fooled by Randomness. And yes, I've seen a lot of claims of various other successes of Taleb, with a lack of any actual citations or proof at all: His own fund folded after yielding mediocre returns, and then he started consulting for a fund that seems to demonstrate a tremendous amount of bl…

I think that you may be missing the point of Fooled by Randomness. The point is knowing whether your success came from luck, skill, or both. It also shined a light on people who took high-probability, "sure bets" (selling out-of-the-money options) that paid off most of the time but would eventually bankrupt them if the stock price jumped.

Quite contrary, the point of the book is that it is very difficult to differentiate between skill and luck when random odds influence events.

For instance if you asked 1000 people to guess coin flips, on average one half will be wrong on each flip. So on the first flip you'd drop to 500, then 250, then 125, then 65, then 32, then 16, then 8, then 4, then 2, then you'd -- in a perfectly ideal scenario -- have one person left. A person that remarkably, "against all odds", guessed 10 coin flips in a row right! Surely they must be some sort of magician or seer, right? Yet their probability of guessing the next coin flip is no better than the people eliminated in the first round.

But that is exactly what we do with things like exceptional events: We find the person who was right about a set of events, among a collection of people guessing almost everything, and assume they've cracked the code. Even outside of financial situations (like being a guy who happened to have a portfolio that did great during an extreme black swan on October 19th, 1987, even if that same portfolio generally did terribly), just look at what happened with 9/11: Of all of the millions of scenarios that people concocted for fiction or just postulating, anyone who talked about a plane hitting the WTC suddenly became prophetic.

It's actually a pretty good book, as an aside.

Re: Nassim Taleb: We should retire the notion of standard deviation

#164

Post by tptacek »

brohee wrote: post #129
a8da6b0c91d wrote: post #44

Earlier quoted context omitted.

Eric Falkenstein's various Taleb takedowns are fun. http://falkenblog.blogspot.com/2009/03/fama-on-taleb.html http://falkenblog.blogspot.com/2012/11/taleb-mishandles-frag... http://falkenblog.blogspot.com/2010/12/nassim-taleb-imitates... http://falkenblog.blogspot.com/2009/03/review-of-talebs-blac...

Is it telling that in his review of Black Swann he's seriously citing Malcolm Gadwell of all people?

No. He draws no conclusions from Gladwell's analysis; he brings Gladwell into the discussion only to show a well-publicized comparison between what Gladwell believed to be two opposing views of how risk works in the markets. He then tries to demonstrate that Niederhoffer's approach outperformed Taleb's, not because Gladwell said so, but because the empirically available facts of how the funds operated appear to demonstrate it.

Gladwell plays into the comparison about as much as the editors of Cosmopolitan would play into it if Niederhoffer and Taleb had shared the cover of an issue of Cosmo.

Re: Nassim Taleb: We should retire the notion of standard deviation

#165

Post by avenger123 »

bluecalm wrote: post #147
tsax wrote: post #133

Earlier quoted context omitted.

Yeah a polyglot who writes classical Greek, Arabic, French, English and can do advanced statistical modeling is clearly an idiot right?

Well, it surely looks like it seeing level of argument he presents and lack of logical consistency in his writing. Even in this article he claims SD doesn't model real world as well as MAD and then as arguments ask you what would you do if you were asked to calculate... MAD, you would clearly see it's not SD :-) His latest book is worse than the article and my conclusion about him being and idiot is based mainly on i…

Although its unfair to nitpick one paragraph since its out of context from the rest of the chapter/book, reading the aging paragraph makes me see where you are coming from with all this.

The guy can't help himself. It's like reading a stock ticker (except its Taleb's stream of thought), flashing across with all the different thoughts that don't necessarily correlate with each other. You think there's some relevancy there but its hard to pick it out in the moment.

I will say though, Taleb reader's are surely great Words with Friends or Scrabble players. You just can't help picking up a few new words.

I don't agree with your 'vile idiot' statement but I mostly concur with your thoughts. I hope that you don't let Taleb affect your senescence.

Re: Nassim Taleb: We should retire the notion of standard deviation

#166

Post by ThomPete »

bluecalm wrote: post #71
vidarh wrote: post #60

Earlier quoted context omitted.

> Funny I find your comments to be doing exactly what they are critiquing. I was about to write the exact same thing. It is very hard to take this criticism seriously when it is written that way.

Replying to you and parents: > Funny I find your comments to be doing exactly what they are critiquing. I am really not doing it. I gave specific examples of what is wrong with his article and in what areas standard deviation works. I am not defending a view that "standard deviation describes reality better" or anything like that. I am saying why his article is bad and in what areas his solution of just using MAD doe…

"...common use..."

I don't think it gets much clearer than that.

Re: Nassim Taleb: We should retire the notion of standard deviation

#167

Post by bravura »

bluecalm wrote: post #16

So first about the article: >>The notion of standard deviation has confused hordes of scientists What an assertion! It also proved to be very useful for hordes of scientists... what about some examples of confused scientists ? >>There is no scientific reason to use it in statistical investigations in the age of the computer As someone who uses it daily I am eagerly awaiting his argument. >>Say someone just asked you…

>>>There is no scientific reason to use it in statistical investigations in the age of the computer

>> As someone who uses it daily I am eagerly awaiting his argument.

Compute-intensive statistical tests, which are data-driven and make fewer assumptions about the underlying distribution, can give tighter confidence intervals and detect statistical significance better.

For example, stratified shuffling.

Re: Nassim Taleb: We should retire the notion of standard deviation

#168

Post by ThomPete »

bluecalm wrote: post #80
ThomPete wrote: post #63

Earlier quoted context omitted.

It might not be interesting to you and I am sure you know how to apply them. It is however important if large groups of disciplines are not able to apply it properly. Instead of calling him a vile idiot and refusing to even look at what he is trying to say you could see this statement as a cry for help, a provocation to start a debate and so on. The deeper issues here, I think is that he is probably not death wrong n…

>It might not be interesting to you and I am sure you know how to apply them. I am saying that pointing out that some people misuse some tools is not interesting without specific examples and problems it causes. >It is however important if large groups of disciplines are not able to apply it properly. Of course it is, let's try to identify such situations... o just let's try to find one problematic example. >It might…

If he doesn't have anything specific to say one would think you wouldn't have anything specific to critique him for.

And you are not the only one who read his books. But again. His criticism aligns with my experience, so what do I know.

Re: Nassim Taleb: We should retire the notion of standard deviation

#169

Post by avenger123 »

edw519 wrote: post #102
bluecalm wrote: post #16

So first about the article: >>The notion of standard deviation has confused hordes of scientists What an assertion! It also proved to be very useful for hordes of scientists... what about some examples of confused scientists ? >>There is no scientific reason to use it in statistical investigations in the age of the computer As someone who uses it daily I am eagerly awaiting his argument. >>Say someone just asked you…

Counterexample: | .. | . . | waste | . . | stuff | of | . . | I | my | . . | don't | time | . stuff I'll . | grok | | . read . | |(Nassim Taleb) | | . | . | | . |______________________________________________________ -3 -2 -1 0 1 2 3 standard deviations

Very nice! I did have his books on my reading list. But this whole discussion definitely puts me closer to the "waste of my time" side of the distribution.

Re: Nassim Taleb: We should retire the notion of standard deviation

#170

Post by tdees40 »

regularfry wrote: post #30
n00b101 wrote: post #15

Taleb has a good point about people mistakenly interpreting standard deviation (sigma) as Mean Absolute Deviation (MAD). I like that he gives some conversions (sigma ~= 1.25 * MAD, for Normal distribution). I think it's rather silly to talk about "retiring" standard deviation, but we can't blame Taleb - the publication itself posed the question "2014: What Scientific Idea is Ready for Retirement?" to various scientif…

I think his point is that Black-Scholes et al are holed beneath the waterline precisely because they involve standard deviations. In his world, you're better off being unable to price an option than you would be with Black-Scholes. Your example of "That put option is currently selling at 30% vol" is actually an example of why the system is so completely broken: if volatility as standard deviation was valid, all optio…

I've never understood this part of Taleb's argument. Of course, constant vol Black-Scholes does not hold. BUT NO ONE USES THIS. Everyone in the financial industry is well aware of the volatility skew, and spends lots of time adjusting for it.

B-S vols are putting the "wrong number into the wrong equation to get the right price" as Rebonnato famously said.

Post reply on HN