Misbehaviour of markets and the Black Swan

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
User avatar
Topic Author
305pelusa
Posts: 289
Joined: Fri Nov 16, 2018 10:20 pm

Misbehaviour of markets and the Black Swan

Post by 305pelusa » Sun Jun 30, 2019 5:05 pm

Just read these two books by Mandelbrot and Nassim and I'm absolutely fascinated by the material. I always kinda knew asset prices didn't follow a neat Bell Curve and that they displayed fat tails. These books go far beyond that though. That the distributions aren't Gaussian at all, and instead display wild variation (the standard deviation itself constantly changes) and follow the Power Law.

The implications are almost nihilistic in a way. Standard deviation doesn't just underestimate risk; it's totally innapropiate as a measure of risk. Correlation doesn't necessitate normality but it does require a finite standard deviation so that measure is also completely useless.

It always felt strange to use St. Dev and correlations to make portfolios when they are so widely varying depending on the period you measure them. Now I understand why. It seems to me that they are perhaps innapropiate for these distributions.

The conclusion has huge consequences on the application of MPT.

Anyone one else read these books and perhaps wants to talk about how they have changed their thinking with their finances? What action items could come out if you believe in they're material?

livesoft
Posts: 66960
Joined: Thu Mar 01, 2007 8:00 pm

Re: Misbehaviour of markets and the Black Swan

Post by livesoft » Sun Jun 30, 2019 5:11 pm

There are other books that cover this subject. I think you need to read "The Physics of Wall Street" by James Owen Weatherall who starts back with Louis Bachelier in the 1890s and goes through a few people over the years that keep rediscovering things. He does cover Mandelbrot, too. You will learn about a few different kinds of distributions that are not Gaussian. Have fun!

Guess what? I came away believing not in the physics, but in the behavioral economics side of things. Some people would consider me a physicist, too. So I use a technique that helps me with all this.
Wiki This signature message sponsored by sscritic: Learn to fish.

User avatar
nisiprius
Advisory Board
Posts: 38471
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: Misbehaviour of markets and the Black Swan

Post by nisiprius » Sun Jun 30, 2019 5:37 pm

Yes. The Mandelbrot book in particular amazed me. I felt as if the scales had fallen from my eyes. I think it is interesting how rarely mainstream investment writers and experts talk about it. "Nihilistic" is exactly the word I would use, because if Mandelbrot is right, 99% of the financial economics used in investing is nonsense. Furthermore, I don't think the mainstream has any answer to it. Their response is just to look away and pretend it isn't there.

The word "obscene" is literally derived from words meaning "look away" or "avert your eyes," something so repellent you can't bear to look at it. To the investing industry, The Misbehavior of Markets is an obscene book.

Fooled by Randomness and The Black Swan were discussed quite a lot in this forum when they came out, and "black swan" has become part of the language. I'm about halfway through Taleb's book, Skin in the Game. I have mixed feelings. I feel that his writing does drift off into self-indulgence and unbridled ego. I find his books stimulating to read and full of interesting ideas, but I don't put him in the same class as Mandelbrot.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.

User avatar
Topic Author
305pelusa
Posts: 289
Joined: Fri Nov 16, 2018 10:20 pm

Re: Misbehaviour of markets and the Black Swan

Post by 305pelusa » Sun Jun 30, 2019 5:53 pm

nisiprius wrote:
Sun Jun 30, 2019 5:37 pm
Yes. The Mandelbrot book in particular amazed me. I felt as if the scales had fallen from my eyes. I think it is interesting how rarely mainstream investment writers and experts talk about it. "Nihilistic" is exactly the word I would use, because if Mandelbrot is right, 99% of the financial economics used in investing is nonsense. Furthermore, I don't think the mainstream has any answer to it. Their response is just to look away and pretend it isn't there.

The word "obscene" is literally derived from words meaning "look away" or "avert your eyes," something so repellent you can't bear to look at it. To the investing industry, The Misbehavior of Markets is an obscene book.

Fooled by Randomness and The Black Swan were discussed quite a lot in this forum when they came out, and "black swan" has become part of the language. I'm about halfway through Taleb's book, Skin in the Game. I have mixed feelings. I feel that his writing does drift off into self-indulgence and unbridled ego. I find his books stimulating to read and full of interesting ideas, but I don't put him in the same class as Mandelbrot.
Agree 100% with all of this. It essentially invalidates everything I've read about on portfolio theory.

I actually really enjoy Nassim's style and I have to admit that it drilled on me just how relevant Mandelbrot's work really is. And for some reason, I don't mind the ego and self-indulgence. I'm not sure I'd like him to speak to me directly that way but because it's a book and it's more like I'm reading his thoughts, I like it as unfiltered as it is. Overall, I just don't really mind it.


livesoft wrote:
Sun Jun 30, 2019 5:11 pm
There are other books that cover this subject. I think you need to read "The Physics of Wall Street" by James Owen Weatherall who starts back with Louis Bachelier in the 1890s and goes through a few people over the years that keep rediscovering things. He does cover Mandelbrot, too. You will learn about a few different kinds of distributions that are not Gaussian. Have fun!

Guess what? I came away believing not in the physics, but in the behavioral economics side of things. Some people would consider me a physicist, too. So I use a technique that helps me with all this.
I will read that next, thanks!

bgf
Posts: 959
Joined: Fri Nov 10, 2017 9:35 am

Re: Misbehaviour of markets and the Black Swan

Post by bgf » Sun Jun 30, 2019 6:25 pm

OP, i too just finished misbehavior of markets and read taleb's books a while back. i can also vouch for the physics of wall street, which i really enjoyed.

I'd highly recommend 'a man for all markets' by ed thorp! i think you'll really like it. at the end of the day, i like his conception of markets the most.

optimization is a fool's errand, yet the amount of time spent handwringing over it here and in academia is mindboggling.

'when genius failed' by lowenstein is also quite good.

if you dont have an articulable, quantifiable edge over the market that can be implemented in real markets, then buy an index fund. period.

not a value fund. not a factor fund. not an active fund. just a cheap index fund.

as for the book, one thing i wish mandelbrot explained a little more was exactly what his definition of risk was. it seemed to me that he defined risk the same way that modern financial theory does, risk = volatility. where mandelbrot differs is that he introduces his idea of "wild randomness." but still, his conception of risk inherent in financial markets appears to be the volatility of market prices, just with the recognition that they are far more volatile that the bell curve suggests, and subject to discontinuity and long term memory.

kind of related to that, and something i also wish he spent more time discussing, is the actual practical scaling of markets over time. i think he said something that basically from hours to a span of a couple years, markets are self-similar, but when you get down into very small time increments, or long time increments, eg. longer than two years, the self-similarity breaks down.

for me, as a long term investor, i wish he'd spent more time discussing long term holding periods. in fairness, that's not what he has been concerned with.
“TE OCCIDERE POSSUNT SED TE EDERE NON POSSUNT NEFAS EST"

alex_686
Posts: 4694
Joined: Mon Feb 09, 2015 2:39 pm

Re: Misbehaviour of markets and the Black Swan

Post by alex_686 » Sun Jun 30, 2019 7:00 pm

So, let me modestly take the other side.

When it comes to standard deviations, I am a "glass half full" type of guy - or to be precise, 95% full. It provides a robust framework, independent of asset time and time horizon. It is simple and robust. And it is good out to 2 to 3 standard deviations. And the defects and shortcomings are well known.

I like Mandelbrot. However it adds lots of complexity and sophistication for very little gain. Plus you loss a lot of universality.

I dislike Nassim. His criticism are valid but were widely known. The problem is that he does not offer much in the way of solutions.

bgf
Posts: 959
Joined: Fri Nov 10, 2017 9:35 am

Re: Misbehaviour of markets and the Black Swan

Post by bgf » Sun Jun 30, 2019 7:37 pm

alex_686 wrote:
Sun Jun 30, 2019 7:00 pm
So, let me modestly take the other side.

When it comes to standard deviations, I am a "glass half full" type of guy - or to be precise, 95% full. It provides a robust framework, independent of asset time and time horizon. It is simple and robust. And it is good out to 2 to 3 standard deviations. And the defects and shortcomings are well known.

I like Mandelbrot. However it adds lots of complexity and sophistication for very little gain. Plus you loss a lot of universality.

I dislike Nassim. His criticism are valid but were widely known. The problem is that he does not offer much in the way of solutions.
i think it is unfair to criticize taleb for "not providing solutions." if i understand him correctly, one of his main points is that there IS NO SOLUTION to financial markets. financial markets are not simple math problems, i.e., portfolios cannot be optimized. he rails against modern financial theory for convincing the financial world that there IS A SOLUTION. the belief that we know the solution has led to many blow ups. this is his point.

you say what he says is well known, and i agree, but the criticism is not well heeded.
“TE OCCIDERE POSSUNT SED TE EDERE NON POSSUNT NEFAS EST"

User avatar
hdas
Posts: 955
Joined: Thu Jun 11, 2015 8:24 am

Re: Misbehaviour of markets and the Black Swan

Post by hdas » Sun Jun 30, 2019 8:24 pm

305pelusa wrote:
Sun Jun 30, 2019 5:05 pm
Anyone one else read these books and perhaps wants to talk about how they have changed their thinking with their finances? What action items could come out if you believe in they're material?
The genius of Taleb was translating options pit wisdom to the masses for entertainment and pseudo philosophy. That’s it. When it comes to making money from trading, he’s utterly useless as evidenced by the failure of his hedge fund venture. He got paid on 87 on an Eurodollar trade and likes to call himself a trader ever since. He’s very good at selling books tho. :greedy
"whenever there is a randomized way of doing something, then there is a nonrandomized way that delivers better performance but requires more thought" ET Jaynes

User avatar
Topic Author
305pelusa
Posts: 289
Joined: Fri Nov 16, 2018 10:20 pm

Re: Misbehaviour of markets and the Black Swan

Post by 305pelusa » Sun Jun 30, 2019 8:56 pm

bgf wrote:
Sun Jun 30, 2019 6:25 pm
OP, i too just finished misbehavior of markets and read taleb's books a while back. i can also vouch for the physics of wall street, which i really enjoyed.

I'd highly recommend 'a man for all markets' by ed thorp! i think you'll really like it. at the end of the day, i like his conception of markets the most.

optimization is a fool's errand, yet the amount of time spent handwringing over it here and in academia is mindboggling.

'when genius failed' by lowenstein is also quite good.

if you dont have an articulable, quantifiable edge over the market that can be implemented in real markets, then buy an index fund. period.

not a value fund. not a factor fund. not an active fund. just a cheap index fund.
Thanks for the book recommendations. I think I'll continue with Antifragile for now but will look into those soon.

Also, I don't follow why the index market fund is the logical conclusion of the topic. If I'm understanding the material right, the mere notion that the market portfolio is the most efficient (even the whole idea of an efficient frontier) goes right out the window.
bgf wrote:
Sun Jun 30, 2019 6:25 pm
as for the book, one thing i wish mandelbrot explained a little more was exactly what his definition of risk was. it seemed to me that he defined risk the same way that modern financial theory does, risk = volatility. where mandelbrot differs is that he introduces his idea of "wild randomness." but still, his conception of risk inherent in financial markets appears to be the volatility of market prices, just with the recognition that they are far more volatile that the bell curve suggests, and subject to discontinuity and long term memory.
I thought he meant that risk, as measured in st. dev. or volatility was out of the window. Instead, it is replaced by a different model altogether, with parameters like alpha. Alpha especially would explain just how likely big events really are, since it is used to describe the exponent of the power law itself.

As Nassim writes though, estimating the alpha is extremely difficult. So for now, I'm not entirely sure how to go about figuring out the possibility of Grey Swans with the Mandelbrotian model. But I'm somewhat certain it's related to the exponent of the power law.

Perhaps I interpreted the text differently.
hdas wrote:
Sun Jun 30, 2019 8:24 pm
305pelusa wrote:
Sun Jun 30, 2019 5:05 pm
Anyone one else read these books and perhaps wants to talk about how they have changed their thinking with their finances? What action items could come out if you believe in they're material?
The genius of Taleb was translating options pit wisdom to the masses for entertainment and pseudo philosophy. That’s it. When it comes to making money from trading, he’s utterly useless as evidenced by the failure of his hedge fund venture. He got paid on 87 on an Eurodollar trade and likes to call himself a trader ever since. He’s very good at selling books tho. :greedy
I can't speak to his hedge fund experience. All I found online is that the fund got 60% returns on 2000, then "anemic" returns for the three years after, and then closed down. Certainly doesn't seem like a failure but I don't have the details of course.

Either way, that his hedge fund venture went poorly means absolutely nothing to me as far as his message and ideas are concerned. I wouldn't trust him any more/less if he had been successful. In fact, that's the common theme of the books. That success (and failure) is far more luck-dependent than we want to believe. If you take that advice for what it's worth, I think it's insightful.

columbia
Posts: 1648
Joined: Tue Aug 27, 2013 5:30 am

Re: Misbehaviour of markets and the Black Swan

Post by columbia » Mon Jul 01, 2019 4:41 am

What are actionable aspects of these books for mere mortals (who should have nothing to do with options and leverage)?

livesoft
Posts: 66960
Joined: Thu Mar 01, 2007 8:00 pm

Re: Misbehaviour of markets and the Black Swan

Post by livesoft » Mon Jul 01, 2019 4:46 am

columbia wrote:
Mon Jul 01, 2019 4:41 am
What are actionable aspects of these books for mere mortals (who should have nothing to do with options and leverage)?
I don't think there are any. But I suppose I could stretch and say the one actionable item is to not to try to compete with these people and just invest in index funds.
Wiki This signature message sponsored by sscritic: Learn to fish.

larryswedroe
Posts: 15942
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: Misbehaviour of markets and the Black Swan

Post by larryswedroe » Mon Jul 01, 2019 8:05 am

FWIW, I have very different view Some key thoughts
First, this idea that markets not normally distributed is OLD news. Fama in fact wrote about this LONG ago. But that is true for daily or very short returns. Get out to a year and they are reasonably close to being log normally distributed. And no one should be equity investor if horizon is very short like days, weeks or even months. Good example, 2008 was worst crisis in post WW 11 era and it was well within the bottom 5% of the Monte Carlo Simulations we ran for clients. So investors should have "expected" it, though not knowing when it could happen, and have built portfolios that they could live with through such a period.

Second, re correlations. It is of course also well known that correlations are not static, just AVERAGES over long periods. Thus, it is important when investing and constructing portfolios to understand not just average correlation but also when correlations are likely to rise and fall.

Third, the important insights from the books mentioned is we cannot know the future and the unexpected does happen and thus should be planned for ---and investors should not take more risk than have the ability, willingness or need to take. The reason is that risks will eventually show up. So they have to be planned for. And when you have uncertainty the safest port in that sea of uncertainty is diversifying across as many unique sources of risk as one can identify that meet all your established criteria for investment---which means accepting being uncomfortable (living with tracking variance in return for benefits of diversification)

BTW, I would add the aforementioned The Physics of Wall Street as important book, it's on my list of best bakers dozen finance books.

Hope that is helpful.
Larry

bgf
Posts: 959
Joined: Fri Nov 10, 2017 9:35 am

Re: Misbehaviour of markets and the Black Swan

Post by bgf » Mon Jul 01, 2019 9:29 am

305pelusa wrote:
Sun Jun 30, 2019 8:56 pm

Also, I don't follow why the index market fund is the logical conclusion of the topic. If I'm understanding the material right, the mere notion that the market portfolio is the most efficient (even the whole idea of an efficient frontier) goes right out the window.

I thought he meant that risk, as measured in st. dev. or volatility was out of the window. Instead, it is replaced by a different model altogether, with parameters like alpha. Alpha especially would explain just how likely big events really are, since it is used to describe the exponent of the power law itself.

As Nassim writes though, estimating the alpha is extremely difficult. So for now, I'm not entirely sure how to go about figuring out the possibility of Grey Swans with the Mandelbrotian model. But I'm somewhat certain it's related to the exponent of the power law.

Perhaps I interpreted the text differently.
1) my conclusion to use index funds isn't just drawn from misbehavior of markets but from all the reading that i've done, which includes that book. also, an indexing strategy can be supported by citing efficient markets, but you can formulate an argument for using indexes without assuming that markets are efficient.

2) with respect to risk, we read it the same way. you are correct that mandelbrot does not use standard deviation as a measure of variance of stock prices. my point was that his fundamental understanding of risk is still grounded in the fluctuation stock prices. he just shows that the traditional way to model those fluctuations is not representative of real markets.

in contrast, others define risk without reference to 'short term' stock price variance. buffett, for example, pays little if any attention to the weekly/monthly meandering of stock prices, but he definitely understands risk.
“TE OCCIDERE POSSUNT SED TE EDERE NON POSSUNT NEFAS EST"

bgf
Posts: 959
Joined: Fri Nov 10, 2017 9:35 am

Re: Misbehaviour of markets and the Black Swan

Post by bgf » Mon Jul 01, 2019 9:31 am

livesoft wrote:
Mon Jul 01, 2019 4:46 am
columbia wrote:
Mon Jul 01, 2019 4:41 am
What are actionable aspects of these books for mere mortals (who should have nothing to do with options and leverage)?
I don't think there are any. But I suppose I could stretch and say the one actionable item is to not to try to compete with these people and just invest in index funds.
if modern portfolio theory is actionable, then so is any book that reveals the flaws and limitations of it.
“TE OCCIDERE POSSUNT SED TE EDERE NON POSSUNT NEFAS EST"

User avatar
bobcat2
Posts: 5337
Joined: Tue Feb 20, 2007 3:27 pm
Location: just barely Outside the Beltway

Re: Misbehaviour of markets and the Black Swan

Post by bobcat2 » Mon Jul 01, 2019 12:30 pm

larryswedroe wrote:
Mon Jul 01, 2019 8:05 am
[T]his idea that markets not normally distributed is OLD news. Fama in fact wrote about this LONG ago. But that is true for daily or very short returns. Get out to a year and they are reasonably close to being log normally distributed. And no one should be equity investor if horizon is very short like days, weeks or even months. Good example, 2008 was worst crisis in post WW 11 era and it was well within the bottom 5% of the Monte Carlo Simulations we ran for clients. So investors should have "expected" it, though not knowing when it could happen, and have built portfolios that they could live with through such a period.


It's true that high frequency stock returns (daily or higher frequency) have very odd distributions to the point that it is unclear exactly what distribution they are following. But for lower frequency returns, such as annual returns, we can appeal to the Central Limit Theorem. Here is a post I made a few years ago that addresses this issue. With thanks to Grabiner for clearly making the essential argument.
grabiner wrote:Note that this is a distribution of daily returns. With relatively low correlation between daily returns, the returns over longer periods are closer to a normal distribution.

This is a general principle in statistics. The Central Limit Theorem says that under certain conditions (independent identical distributions, for example), you can prove that the sum of a large number of events will have a nearly-normal distribution. But even when the conditions do not hold, you expect a normal distribution in practice when you add many small random variables to get a large one, such as 250 daily stock returns to get an annual return.
Hi David,

Daily returns do have much fatter tails than a normal distribution and also exhibit considerable volatility clustering, which is not an attribute of a normal distribution. Annual returns are typically fairly well represented by the log-normal distribution and exhibit little volatility clustering.
There are two principal reasons that annual returns are better represented by the log-normal distribution rather than the normal distribution. Namely, the compounding of the returns and the zero lower bound on stock returns.

Annual returns are slightly more fat tailed than a log-normal distribution but that can be handled by something as simple as applying a GARCH(1,1) model to the log-normal. Sometimes followers of Mandelbrot claim that the Central Limit Theorem doesn't hold in the case of annual stock returns because the returns are not independent of one another. But high frequency returns don't have to be strictly independent for the annual returns to be close to log-normal. So whether the annual returns are approximately log-normal becomes an empirical question. On the grounds of observation and experience, aka empirically, Mandelbrot and his followers appear to be wrong.

If you are familiar with statistics reading a Taleb book is a rather tedious rehash of some aspects of elementary statistics. Although, to give the devil his due, he does do a good job of hammering home some important points that often aren't stressed enough in introductory statistics courses. I can imagine that for some without much prior exposure to statistics reading Taleb would seem to be a revelatory experience.

Taleb is good at creating controversy, self-promotion, and selling books. Somehow I feel that there is possibly another public figure I am familiar with who has those qualities. :wink:

BobK
Last edited by bobcat2 on Mon Jul 01, 2019 12:50 pm, edited 1 time in total.
In finance risk is defined as uncertainty that is consequential (nontrivial). | The two main methods of dealing with financial risk are the matching of assets to goals & diversifying.

Dottie57
Posts: 6361
Joined: Thu May 19, 2016 5:43 pm

Re: Misbehaviour of markets and the Black Swan

Post by Dottie57 » Mon Jul 01, 2019 12:38 pm

livesoft wrote:
Mon Jul 01, 2019 4:46 am
columbia wrote:
Mon Jul 01, 2019 4:41 am
What are actionable aspects of these books for mere mortals (who should have nothing to do with options and leverage)?
I don't think there are any. But I suppose I could stretch and say the one actionable item is to not to try to compete with these people and just invest in index funds.
Done.

User avatar
vineviz
Posts: 4405
Joined: Tue May 15, 2018 1:55 pm

Re: Misbehaviour of markets and the Black Swan

Post by vineviz » Mon Jul 01, 2019 12:44 pm

bobcat2 wrote:
Mon Jul 01, 2019 12:30 pm
I can imagine that for some without much prior exposure to statistics reading Taleb would seem to be a revelatory experience.

Taleb is good at creating controversy, self-promotion, and selling books.

Everything BobK wrote is excellent, but IMHO these two sentences especially are deserving of a hearty "+ 1".
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch

User avatar
nisiprius
Advisory Board
Posts: 38471
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: Misbehaviour of markets and the Black Swan

Post by nisiprius » Mon Jul 01, 2019 2:12 pm

See Larry Swedroe's post immediately below this one for his rebuttal.
larryswedroe wrote:
Mon Jul 01, 2019 8:05 am
...Second, re correlations. It is of course also well known that correlations are not static, just AVERAGES over long periods. Thus, it is important when investing and constructing portfolios to understand not just average correlation but also when correlations are likely to rise and fall...
I understand you to be saying that 1) a expert practitioner can predict future changes in correlations, and that 2) it is "important" to make tactical adjustments based on those anticipated changes.

It seems to me it follows from that, that it would be an important mistake to adopt a strategy that is founded on correlation values, unless one is an expert practitioner with knowledge and skill to predict correlation changes and make tactical adjustments--or is the client of such a practitioner.
Last edited by nisiprius on Mon Jul 01, 2019 9:37 pm, edited 1 time in total.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.

larryswedroe
Posts: 15942
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: Misbehaviour of markets and the Black Swan

Post by larryswedroe » Mon Jul 01, 2019 3:02 pm

Nisiprius
I literally have no clue how you could draw those conclusions when I said nothing like that at all.

In fact I know you cannot have missed my saying numerous times that my crystal ball is always cloudy. So that addresses the issue of tactical changes based on anticipating the events that cause correlations to change. There are no good forecasters who can do that.

What I said is that investors need to acknowledge and understand that correlations are time varying, but that doesn't make them useless as some of the posters were implying. It means you have to understand when they tend to rise (perhaps in bad times for risky assets) and when they tend to fall (perhaps for safe assets relative to risky assets). Knowing that helps you to determine both if you should allocate to them and from where to take the allocation (from stocks or bonds).

Which then throws out your last statement.

I hope that clarifies things
Larry

freyj6
Posts: 216
Joined: Thu Jun 19, 2014 2:09 am

Re: Misbehaviour of markets and the Black Swan

Post by freyj6 » Mon Jul 01, 2019 3:54 pm

@Larry -BTW, I would add the aforementioned The Physics of Wall Street as important book, it's on my list of best bakers dozen finance books.
I'd be really curious to know some of the others, especially the ones that aren't on the bogleheads frequently recommended list.

================

On topic:

I recently read Misbehavior of Markets and I read Taleb's looks a while ago. It does seem as though most of popular and even academic finance is based on faulty assumptions.

I think that's why I loved Bill Bernstein's books so much. I'd been reading several complex, academic books on retirement planning and understanding distribution portfolios, then I revisited Ages of the Investor (a 55 page booklet) and it explained the important things better than any of them.

In my opinion, the key takeaway of these kinds of books is that we need to act as intelligently as possible in the face of uncertainty rather than making fragile assumptions and then supporting them with a gargantuan amount of math. As an analogy, no one knows the perfect diet for humans but we can be almost certain that whole foods are healthier than heavily processed ones, and that a strawberry is healthier than a Twinky. Likewise, we finance we can still make a lot of intelligence decisions, but at all costs we should avoid trying to make those decisions under the assumption that we understand more than we actually do.

ajjulee
Posts: 70
Joined: Tue Jun 17, 2014 2:46 pm

Re: Misbehaviour of markets and the Black Swan

Post by ajjulee » Mon Jul 01, 2019 4:22 pm

bgf wrote:
Sun Jun 30, 2019 7:37 pm
alex_686 wrote:
Sun Jun 30, 2019 7:00 pm
So, let me modestly take the other side.

When it comes to standard deviations, I am a "glass half full" type of guy - or to be precise, 95% full. It provides a robust framework, independent of asset time and time horizon. It is simple and robust. And it is good out to 2 to 3 standard deviations. And the defects and shortcomings are well known.

I like Mandelbrot. However it adds lots of complexity and sophistication for very little gain. Plus you loss a lot of universality.

I dislike Nassim. His criticism are valid but were widely known. The problem is that he does not offer much in the way of solutions.
i think it is unfair to criticize taleb for "not providing solutions." if i understand him correctly, one of his main points is that there IS NO SOLUTION to financial markets. financial markets are not simple math problems, i.e., portfolios cannot be optimized. he rails against modern financial theory for convincing the financial world that there IS A SOLUTION. the belief that we know the solution has led to many blow ups. this is his point.

you say what he says is well known, and i agree, but the criticism is not well heeded.
I'm not a huge fan of Taleb either. If one were to put a gun to my head, I could probably explain in a lengthy passage why I don't like his writing that will bore everyone including my mother (after making sure there are real bullets in the gun and that the gun holder knew how to fire a gun etc.,) but I have one comment on his big idea of 'Black Swan' - that which cannot be foreseen and hence no way to prepare for it. Do such events really exist in life? What are those? Aliens attacking earthlings? Men giving birth to children? Terrorists attack on major land marks (or Aliens attacking major landmarks such as White House, Capital Records Tower in LA etc., as an aside, aren't they better off attacking power grids or water supplies?), Subprime lending? US president who is in the pay of a foreign enemy government? All these have been discussed or written about long before such events happened or may happen in future. Black Swan, literally, isn't 'so out there' concept either. Everyone knows swans, black is just the opposite of white and there is that ugly duckling...Just thinking :)
What I write on this forum is sometimes the truth, maybe the whole truth, and not always nothing but the truth, for the purposes of online anonymity.

bgf
Posts: 959
Joined: Fri Nov 10, 2017 9:35 am

Re: Misbehaviour of markets and the Black Swan

Post by bgf » Mon Jul 01, 2019 4:40 pm

vineviz wrote:
Mon Jul 01, 2019 12:44 pm
bobcat2 wrote:
Mon Jul 01, 2019 12:30 pm
I can imagine that for some without much prior exposure to statistics reading Taleb would seem to be a revelatory experience.

Taleb is good at creating controversy, self-promotion, and selling books.

Everything BobK wrote is excellent, but IMHO these two sentences especially are deserving of a hearty "+ 1".
i also would imagine that mandelbrot had a decent grasp of elementary statistics... what drives me crazy is how dismissive people are of valid criticism.
“TE OCCIDERE POSSUNT SED TE EDERE NON POSSUNT NEFAS EST"

User avatar
vineviz
Posts: 4405
Joined: Tue May 15, 2018 1:55 pm

Re: Misbehaviour of markets and the Black Swan

Post by vineviz » Mon Jul 01, 2019 5:12 pm

bgf wrote:
Mon Jul 01, 2019 4:40 pm
vineviz wrote:
Mon Jul 01, 2019 12:44 pm
bobcat2 wrote:
Mon Jul 01, 2019 12:30 pm
I can imagine that for some without much prior exposure to statistics reading Taleb would seem to be a revelatory experience.

Taleb is good at creating controversy, self-promotion, and selling books.

Everything BobK wrote is excellent, but IMHO these two sentences especially are deserving of a hearty "+ 1".
i also would imagine that mandelbrot had a decent grasp of elementary statistics... what drives me crazy is how dismissive people are of valid criticism.
Mandelbrot was not a dunce, clearly, and I don't think anyone (including me) is guilty of suggesting that he was. He was a well-respected scientist and mathematician.

That said, his study of finance was as an outsider.To say that it was a hobby for him probably undersells his mastery, but he had AFAIK no formal training in economics or finance. As a result, it's not hard to understand why wrote as if he had discovered something that every finance PhD student learned in their first semester of coursework.

It's not that his criticisms (or those of Nassim Taleb) of financial economics are completely invalid. They aren't. They are, however, both overblown (hyperbole sells books, after all) and well-known.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch

User avatar
bobcat2
Posts: 5337
Joined: Tue Feb 20, 2007 3:27 pm
Location: just barely Outside the Beltway

Re: Misbehaviour of markets and the Black Swan

Post by bobcat2 » Mon Jul 01, 2019 5:16 pm

bgf wrote:
Mon Jul 01, 2019 4:40 pm
i also would imagine that mandelbrot had a decent grasp of elementary statistics... what drives me crazy is how dismissive people are of valid criticism.
I most certainly did not write that Mandelbrot didn't understand basic statistics. What I did write about was that Mandelbrot and some of his followers claim the Central Limit Theorem does not apply to annual stock returns and that that claim appears to be invalid. And I explained why it appears to be invalid - both in theory and empirically. Mandelbrot is certainly correct that high frequency stock returns are non-normal (way non normal) and exhibit volatility clustering.

It was with regard to Taleb's books that I wrote about introductory and/or elementary statistics. Here's the paragraph on Taleb and introductory statistics.
If you are familiar with statistics reading a Taleb book is a rather tedious rehash of some aspects of elementary statistics. Although, to give the devil his due, he does do a good job of hammering home some important points that often aren't stressed enough in introductory statistics courses. I can imagine that for some without much prior exposure to statistics reading Taleb would seem to be a revelatory experience.
BobK
In finance risk is defined as uncertainty that is consequential (nontrivial). | The two main methods of dealing with financial risk are the matching of assets to goals & diversifying.

Carol88888
Posts: 168
Joined: Wed Jan 24, 2018 2:24 am

Re: Misbehaviour of markets and the Black Swan

Post by Carol88888 » Mon Jul 01, 2019 5:24 pm

When I first read "Fooled by Randomness" in 2001 it was transformative for me. I had just gotten out of the market with my biggest gains ever but the book smashed any idea that my results were dependent on any skill I had. Instead, I had simply been in a hot sector during a tremendous bubble and had taken the opportunity to book my profits before they evaporated. I was a "lucky fool" in Taleb's words.

But what I never could get comfortable with was the way Taleb actually invests. He seems to be able to withstand death by a thousand cuts waiting and waiting for the eventual big payday. I think this goes against the psychological make-up of most investors which is probably why his hedge fund suffered redemptions and closed.

I would rather take my pain in the normal gyrations of an index fund which I believe will pay off eventually since I believe in the productivity of American markets. This is akin to Warren Buffett's confidence in US markets over the very long term.

I don't believe in totally efficient markets and I don't think Warren Buffett or Charlie Munger do either since their success shows the market can be beaten with a value methodology, correctly applied.

All I need to know is that an index strategy will probably do better than other strategy I could reasonably execute. And that's an easy assumption for me.

bgf
Posts: 959
Joined: Fri Nov 10, 2017 9:35 am

Re: Misbehaviour of markets and the Black Swan

Post by bgf » Mon Jul 01, 2019 5:57 pm

vineviz wrote:
Mon Jul 01, 2019 5:12 pm

Mandelbrot was not a dunce, clearly, and I don't think anyone (including me) is guilty of suggesting that he was. He was a well-respected scientist and mathematician.

That said, his study of finance was as an outsider.To say that it was a hobby for him probably undersells his mastery, but he had AFAIK no formal training in economics or finance. As a result, it's not hard to understand why wrote as if he had discovered something that every finance PhD student learned in their first semester of coursework.

It's not that his criticisms (or those of Nassim Taleb) of financial economics are completely invalid. They aren't. They are, however, both overblown (hyperbole sells books, after all) and well-known.
again, i think this is unfair to mandelbrot. 1) he was fama's thesis adviser. 2) i can see no reason why being an "outsider" would be a criticism. if anything, it would give him a more objective view as his own career didn't hinge on the validity of what he was studying. 3) these things are well known primarily because of mandelbrot.

"Mandelbrot, like Prime Minister Churchill before him, promises us not utopia but blood, sweat, toil and tears. If he is right, almost all of our statistical tools are obsolete -least squares, spectral analysis, workable maximum likelihood solutions, all our established sample theory, closed distribution functions. Almost without exception, past econometric work is meaningless. Surely, before consigning centuries of work to the ash pile, we should like to have some assurance that all our work is truly useless." (Cootner, The random character of stock market prices 1964, p. 337).
“TE OCCIDERE POSSUNT SED TE EDERE NON POSSUNT NEFAS EST"

bgf
Posts: 959
Joined: Fri Nov 10, 2017 9:35 am

Re: Misbehaviour of markets and the Black Swan

Post by bgf » Mon Jul 01, 2019 6:13 pm

bobcat2 wrote:
Mon Jul 01, 2019 5:16 pm
I most certainly did not write that Mandelbrot didn't understand basic statistics. What I did write about was that Mandelbrot and some of his followers claim the Central Limit Theorem does not apply to annual stock returns and that that claim appears to be invalid. And I explained why it appears to be invalid - both in theory and empirically. Mandelbrot is certainly correct that high frequency stock returns are non-normal (way non normal) and exhibit volatility clustering.


BobK
apologies if i mischaracterized what you said. i just get frustrated by how people here take certain things as absolute gospel. while i acknowledge that its only natural for people versed in a field to attack new, contradictory ideas (kuhn wrote a pretty good book about it!) it doesn't make it less exasperating.
“TE OCCIDERE POSSUNT SED TE EDERE NON POSSUNT NEFAS EST"

bgf
Posts: 959
Joined: Fri Nov 10, 2017 9:35 am

Re: Misbehaviour of markets and the Black Swan

Post by bgf » Mon Jul 01, 2019 6:15 pm

Carol88888 wrote:
Mon Jul 01, 2019 5:24 pm

I don't believe in totally efficient markets and I don't think Warren Buffett or Charlie Munger do either since their success shows the market can be beaten with a value methodology, correctly applied.

All I need to know is that an index strategy will probably do better than other strategy I could reasonably execute. And that's an easy assumption for me.
this is pretty much how i look at things as well. to nitpick, i think there is more to what buffett and munger did than execute a value strategy, which around here is at risk of devolving into some factor explanation, e.g., Buffett's Alpha.
“TE OCCIDERE POSSUNT SED TE EDERE NON POSSUNT NEFAS EST"

columbia
Posts: 1648
Joined: Tue Aug 27, 2013 5:30 am

Re: Misbehaviour of markets and the Black Swan

Post by columbia » Mon Jul 01, 2019 6:36 pm

bgf wrote:
Mon Jul 01, 2019 6:15 pm
Carol88888 wrote:
Mon Jul 01, 2019 5:24 pm

I don't believe in totally efficient markets and I don't think Warren Buffett or Charlie Munger do either since their success shows the market can be beaten with a value methodology, correctly applied.

All I need to know is that an index strategy will probably do better than other strategy I could reasonably execute. And that's an easy assumption for me.
this is pretty much how i look at things as well. to nitpick, i think there is more to what buffett and munger did than execute a value strategy, which around here is at risk of devolving into some factor explanation, e.g., Buffett's Alpha.

For one, actually owning companies is far different than owning stock in companies from the comfort of your couch.

User avatar
Topic Author
305pelusa
Posts: 289
Joined: Fri Nov 16, 2018 10:20 pm

Re: Misbehaviour of markets and the Black Swan

Post by 305pelusa » Mon Jul 01, 2019 7:37 pm

larryswedroe wrote:
Mon Jul 01, 2019 3:02 pm
Nisiprius
I literally have no clue how you could draw those conclusions when I said nothing like that at all.

In fact I know you cannot have missed my saying numerous times that my crystal ball is always cloudy. So that addresses the issue of tactical changes based on anticipating the events that cause correlations to change. There are no good forecasters who can do that.

What I said is that investors need to acknowledge and understand that correlations are time varying, but that doesn't make them useless as some of the posters were implying. It means you have to understand when they tend to rise (perhaps in bad times for risky assets) and when they tend to fall (perhaps for safe assets relative to risky assets). Knowing that helps you to determine both if you should allocate to them and from where to take the allocation (from stocks or bonds).

Which then throws out your last statement.

I hope that clarifies things
Larry
Larry,
No, but I think I see Nisiprius' point. If correlations are sometimes positive and sometimes negative, then for you to make use of the information you would need to predict to an extent when that would occur and then make portfolio changes based on that.

The former, you already admit you do to an extent. Unfortunately, I don't feel at all confident in the assumption that correlation changes in any form of predictable way (i.e. rises in bad times or falls in good times). That's just me though.

Now if you understand that correlation changes in different scenarios, the information has limited use if it doesn't prompt you to make a change in asset allocation.

The only conceivable way I can see correlation (and understanding how/when it changes) being useful with a fixed allocation (no tactical changes) is that you design the portfolio with the worst-case positive high correlation in mind.

I get that your process isn't so regimented. Maybe you make a portfolio with average correlation, then stress-test with the worst, etc. That qualitative use of correlation I think might be about right.
But there are other approaches in MPT (and risk-parity) where the correlation number is used as-is. So you have to pick something. Here's where I'm starting to think the utility is essentially zero.
bobcat2 wrote:
Mon Jul 01, 2019 12:30 pm
I can imagine that for some without much prior exposure to statistics reading Taleb would seem to be a revelatory experience.
I'm not really sure what part of Taleb's writing would be revelatory from a statistics perspective. It barely goes into the technical nature of it. Perhaps that's because I read Mandelbrot first.

The part I found revelatory was the social commentary about the human biases he mentions. Yeah I always understood confirmation bias, and survivorship bias, etc. But they way he applies it, interconnected to chaos theory, non-linear systems and ultimately, precisely why it gives birth to the Black Swan was absolutely fantastic. The writing is extremely fragmented and "all over the place", a constant negative I read online. But somehow, it all clicked at once with me. The stability fallacy of refugees, the success of book writing, the scalable random qualities, etc.

And it was right about the time when he writes towards the end of Part 1 about "the reader might realize everything in this Part really just boils down to the exact same thing". And it all hit me like a train. I liked the book overall.

User avatar
vineviz
Posts: 4405
Joined: Tue May 15, 2018 1:55 pm

Re: Misbehaviour of markets and the Black Swan

Post by vineviz » Mon Jul 01, 2019 8:29 pm

bgf wrote:
Mon Jul 01, 2019 5:57 pm
1) he was fama's thesis adviser.
No, he wasn't. Mandelbrot was never on the faculty at University of Chicago. Fama's advisers were William Alberts, Lawrence Fisher, Robert Graves, James Lorie, Merton Miller, Harry Roberts, and Lester Telser. Fama acknowledged Mandelbrot in a note in the published version of this dissertation, but Mandelbrot was not an adviser. In fact, Mandelbrot – in his 1966 paper -credited Fama for helping him synthesize his ideas on the subject.
bgf wrote:
Mon Jul 01, 2019 5:57 pm
2) i can see no reason why being an "outsider" would be a criticism. if anything, it would give him a more objective view as his own career didn't hinge on the validity of what he was studying.
I didn't remark on his outsider status as a criticism, merely as a possible explanation for his unfamiliarity with the knowledge base of the field.
bgf wrote:
Mon Jul 01, 2019 5:57 pm
3) these things are well known primarily because of mandelbrot.
I think this is overly generous, but it is definitely true that Mandelbrot's work in the early 1960s on prices (e.g. "The variation of certain speculative prices") had some influence on the field and were taken seriously by economists. My comments shouldn't be taken to mean I acknowledge NO contributions by Mandelbrot to the field of economics.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch

alex_686
Posts: 4694
Joined: Mon Feb 09, 2015 2:39 pm

Re: Misbehaviour of markets and the Black Swan

Post by alex_686 » Mon Jul 01, 2019 8:35 pm

bgf wrote:
Sun Jun 30, 2019 7:37 pm
i think it is unfair to criticize taleb for "not providing solutions." if i understand him correctly, one of his main points is that there IS NO SOLUTION to financial markets. financial markets are not simple math problems, i.e., portfolios cannot be optimized. he rails against modern financial theory for convincing the financial world that there IS A SOLUTION. the belief that we know the solution has led to many blow ups. this is his point.
I think we are on the same page, and I think it is a fair criticism of Taleb. His main target was not the individual evaluating their portfolio, but rather regulators regulating risk in the financial markets. I personally don't think it is valid that banks regulate themselves or for bank regulators to use heuristics (i.e., their gut feeling). Even at the individual level he is throwing out the baby with the dirty bathwater. Simple tools like standard deviation are, in my opinion, the right level of sophistication for the individual investor. They are not bullet proof, but better then throwing darts at the board.

User avatar
bobcat2
Posts: 5337
Joined: Tue Feb 20, 2007 3:27 pm
Location: just barely Outside the Beltway

Re: Misbehaviour of markets and the Black Swan

Post by bobcat2 » Mon Jul 01, 2019 9:23 pm

Here is a post I made at Bogleheads a few years ago about Mandelbrot, fat tails, and volatility clustering.

Mandlebrot was not the first to note that stock returns exhibit fat tails. He is, however, generally credited as being the first to discover that stock returns exhibit volatility clustering. Here is Mandlebrot writing in 1963 of what we now know as volatility clustering.
"large changes tend to be followed by large changes, of either sign, and small changes tend to be followed by small changes (of either sign)."

The best concise description I can find of the history of researchers studying the twin problems of fat tails and volatility clustering in stock market return data is in Mark Rubinstein’s fine book, A History of the Theory of Investments, pg 185-190.
Mandlebrot (1963) used stable-Paretian distributions to explain the fat tails observed by Osborne (1959) and Alexander (1961) in the frequency distributions of returns of stocks. However, financial economists were reluctant to adapt this model primarily because they would have to give up variance as their favored measure of risk (since the variance of a stable-Paretian random variable is infinite). In the end, the stable-Paretian hypothesis proved a dead end, particularly as alternative finite-variance explanations of stock market returns were developed.
Mark Rubinstein then goes on to discuss early efforts in the 1960’s and 1970’s to address fat tails in return data and notes the following about an early model with finite variance.
Unfortunately, the model has a serious problem. As Praetz himself notes, actual volatility clusters. There are often consecutive periods of abnormally high volatility and other consecutive periods of abnormally low volatility. But Praetz has assumed that the level of variance is random so that unusually high variance today is likely to be followed by normal variance.

Rubenstein then discusses the first model with finite variance to directly address volatililty clustering and indirectly fat tails.
A much better hypothesis is to suppose that the change in the level of variance is an independent and identically distributed random variable. This allows for volatility clustering. The first to propose a model of this sort is Barr Rosenberg (1972). …

While maintaining the assumption that each individual price change is drawn from a normal distribution, Rosenberg argues that a potential explanation for the high kurtosis (fat tails) is nonstationary of the variance of the normal distribution from which the price is drawn. …

Bar Rosenberg then employs a simple model of stochastic volatility: The predicted next month squared price change (the current variance) equals a fixed linear function of the previous 10 squared monthly price changes: …
Rosenberg then estimates this simple regression on 75 years of monthly data on the S&P Composite Stock Index and it works (the slope coefficient has a t score above 10). Rubinstein then goes on to note that Rosenberg’s study may be the first clear statistical evidence for volatility clustering in stock returns.


And here are some of the conclusions from Barr Rosenberg’s unpublished paper, which in retrospect appear prescient.
The results of the experiment have widespread implications for financial management and the theory of security markets.
i. the requirements for forecasts of price variance;
ii. the opening of the study of the determinants of price variance as a field of economic analysis;
iii. the need to respond to fluctuations in variance in portfolio management;
iv. the role of fluctuations in variance, through the effect on the riskiness of investment and, hence, on the appropriate risk premium, as an influence on the price level.
Mark Rubinstein then goes on to observe that a decade later Robert Engle, apparently unaware of Rosenberg’s unpublished paper, wrote a paper describing a more formal econometric model of the statistical procedure Rosenberg employed. Engle (1982) labeled his model an autoregressive conditional heteroscedasticity (or ARCH) model. And here is a final quote from Rubinstein's book.
In 2003, Engle won the Nobel Prize in Economic Science “for methods of analyzing economic time series with time varying volatility (ARCH).”


Here are links to two threads that earlier this year discussed these topics.
http://www.bogleheads.org/forum/viewtop ... le&start=0

http://www.bogleheads.org/forum/viewtop ... ight=engle

In other words economists put most of their research efforts into volatility clustering rather than fat tails, while Mandlebrot continued to try to address the fat tails directly. Mandlebrot's approach has shown very little progress over the last 40 years. Focusing on volatility clustering has resulted in major advances in the last 30 years.

There are several reasons economists have focused on volatility clustering rather than fat tails. While stock returns are random, stock volatility is not random. (Something that clusters cannot be random.) So they focused on the somewhat predictable aspect of the problem. Secondly, a significant amount of the fat tails in the distribution can be explained by the volatility clustering. So if the time varying volatility can be modeled we have also explained most of the fat tail. Thirdly, volatility is the primary measure of risk in financial economics. Mandlebrot's approach uses methods with infinite volatility. So volatility as a measure of risk goes out the window using Mandlebrot's approach.

Finally here is a survey of the development of financial risk management since the early 20th century by financial econometrician Francis Diebold.

Link - https://www.sas.upenn.edu/~fdiebold/pap ... dElgar.pdf

BobK
In finance risk is defined as uncertainty that is consequential (nontrivial). | The two main methods of dealing with financial risk are the matching of assets to goals & diversifying.

bgf
Posts: 959
Joined: Fri Nov 10, 2017 9:35 am

Re: Misbehaviour of markets and the Black Swan

Post by bgf » Tue Jul 02, 2019 7:42 am

vineviz wrote:
Mon Jul 01, 2019 8:29 pm
bgf wrote:
Mon Jul 01, 2019 5:57 pm
1) he was fama's thesis adviser.
No, he wasn't. Mandelbrot was never on the faculty at University of Chicago. Fama's advisers were William Alberts, Lawrence Fisher, Robert Graves, James Lorie, Merton Miller, Harry Roberts, and Lester Telser. Fama acknowledged Mandelbrot in a note in the published version of this dissertation, but Mandelbrot was not an adviser. In fact, Mandelbrot – in his 1966 paper -credited Fama for helping him synthesize his ideas on the subject.
bgf wrote:
Mon Jul 01, 2019 5:57 pm
2) i can see no reason why being an "outsider" would be a criticism. if anything, it would give him a more objective view as his own career didn't hinge on the validity of what he was studying.
I didn't remark on his outsider status as a criticism, merely as a possible explanation for his unfamiliarity with the knowledge base of the field.
bgf wrote:
Mon Jul 01, 2019 5:57 pm
3) these things are well known primarily because of mandelbrot.
I think this is overly generous, but it is definitely true that Mandelbrot's work in the early 1960s on prices (e.g. "The variation of certain speculative prices") had some influence on the field and were taken seriously by economists. My comments shouldn't be taken to mean I acknowledge NO contributions by Mandelbrot to the field of economics.
take it up with fama and mandelbrot. mandelbrot, in his book, states in no uncertain terms that he was Fama's thesis adviser. you are correct though, mandelbrot was not on the faculty. he was with IBM at the time.
“TE OCCIDERE POSSUNT SED TE EDERE NON POSSUNT NEFAS EST"

bgf
Posts: 959
Joined: Fri Nov 10, 2017 9:35 am

Re: Misbehaviour of markets and the Black Swan

Post by bgf » Tue Jul 02, 2019 7:48 am

bobcat2, thank you for taking the time to post. im going to check out rubenstein's book; it sounds very interesting.

i thought Diebold was pretty fair in section 8 discussing K u and U.
“TE OCCIDERE POSSUNT SED TE EDERE NON POSSUNT NEFAS EST"

larryswedroe
Posts: 15942
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: Misbehaviour of markets and the Black Swan

Post by larryswedroe » Tue Jul 02, 2019 7:59 am

305
.
If correlations are sometimes positive and sometimes negative, then for you to make use of the information you would need to predict to an extent when that would occur and then make portfolio changes based on that.
Let me try again because above is incorrect. You don't need to predict anything. Example, we know that correlations of US and international and EM are time varying and tend to rise towards 1 in crises. You don't need to predict when they go up or down to decide to invest internationally, only that you know sometimes the diversification will be more effective than others. And you accept that. But you would not take EM allocation from bonds because correlations can rise at wrong time.

Take treasuries and stocks, the correlation is about 0. But the tendency is during crises the correlations to turn sharply negative, so that is a much better hedge of equity risk than when correlations tend to rise. Reverse true for junk bonds as correlations are fairly low to stocks but correlation tends to rise sharply in bad times. Don't need to predict anything to understand that if going to invest in junk bonds allocation should not come from safe bonds but from equities.

Carry trade is another example---basically no correlation but correlations tend to rise sharply during crisis, flight to quality. So again this helps you decide if you want to invest and where the allocation should come from.

Commodities is another good example, correlations low but can turn sharply negative in negative supply shocks or sharply positive in negative demand shocks. So have to be aware of that before investing and certainly should influence where you take the allocation from, and then if you add them should consider adding duration risk because of how that mixes.

I don't make any tactical moves based on changing correlations, because no one can predict which economic regimes will show up. But that is not needed at all in deciding on the AA.

Hope that is helpful
Larry

User avatar
nisiprius
Advisory Board
Posts: 38471
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: Misbehaviour of markets and the Black Swan

Post by nisiprius » Tue Jul 02, 2019 8:05 am

bgf wrote:
Tue Jul 02, 2019 7:42 am
vineviz wrote:
Mon Jul 01, 2019 8:29 pm
bgf wrote:
Mon Jul 01, 2019 5:57 pm
1) he was fama's thesis adviser.
No, he wasn't. Mandelbrot was never on the faculty at University of Chicago.
Mandelbrot, in his book, states in no uncertain terms that he was Fama's thesis adviser.
The passage is in Chapter III, in the section entitled "The Efficient Market." My Kindle edition which supposedly has real page numbers shows it as being on page 54:
As a student at the University of Chicago, Fama contacted me at IBM and Harvard; I became his thesis adviser, by telephone, mail, and repeated visits. His dissertation was on my view of market dynamics (of which more later). But we often discussed Bachelier's ideas beyond the model of independent increments, and in subsequent years Fama elaborated them into what is now called the Efficient Markets Hypothesis.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.

alex_686
Posts: 4694
Joined: Mon Feb 09, 2015 2:39 pm

Re: Misbehaviour of markets and the Black Swan

Post by alex_686 » Tue Jul 02, 2019 8:11 am

A note on correlations and crisis. Yes, correlations increase during tines of volitility. This might be significant, or maybe not.

Correlation calculations assume constant volatility. If you don’t have constant volatility you are goyto get garbage numbers. I can get a strong correlation with fair dice if I change the volatility.

For example, during the Asian crisis of the 90s, correlations increased due to the stastical artifact of increased volatility, but in retrospect still offered a great diversification benefit.

User avatar
vineviz
Posts: 4405
Joined: Tue May 15, 2018 1:55 pm

Re: Misbehaviour of markets and the Black Swan

Post by vineviz » Tue Jul 02, 2019 8:37 am

bgf wrote:
Tue Jul 02, 2019 7:42 am
mandelbrot, in his book, states in no uncertain terms that he was Fama's thesis adviser.
Given what I know about his personality, this claim doesn't surprise me.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch

User avatar
nisiprius
Advisory Board
Posts: 38471
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: Misbehaviour of markets and the Black Swan

Post by nisiprius » Tue Jul 02, 2019 8:45 am

To me, the key question is "how fast and how reliably do averages and other measures settle down, as the length of the sampling period increases? If you have a sample of past behavior, how much confidence do you have in decisions you can make about future behavior?"

The Cauchy distribution is an example of one in which nothing settles down at all, not even the average, no matter how long the sampling period is. And it's not some crazy Cantor dust thing, it is a very tame-looking distribution that is approximated in the real world by very reasonable physical models. For example, if you imagine a line source of light--like an infinitely long, infinitely thin fluorescent bulb--the pattern of illumination it casts on a flat surface has the Cauchy distribution.

Since the Cauchy distribution has no mean and no higher moments, I don't think samples from a Cauchy distribution would obey the central limit theorem at all.

Mandelbrot shows three curves on page 39: Gaussian, Cauchy, and an intermediate curve which "will serve later in this book to represent the distribution of price increments of cotton." He elsewhere says things that I interpret to mean that real-world financial data is not as bad as Cauchy, but much worse than the normal distribution.

Now, Wikipedia's article on the central limit theorem says:
The central limit theorem gives only an asymptotic distribution. As an approximation for a finite number of observations, it provides a reasonable approximation only when close to the peak of the normal distribution; it requires a very large number of observations to stretch into the tails.
So while the central limit theorem might apply theoretically to financial data, it might well settle down much less slowly. It doesn't do us much good to invoke the central limit theorem if it requires an infinite amount of data, or a thousand years of data, for the distribution of the same average to get close to a normal distribution.

And while I don't have an answer, that is really the crux of the question: how fast do data averages settle down to a normal distribution, and how close do they get? I interpret Mandelbrot as saying that mainstream financial economists are assuming that financial data is tamer than it is; that it does not settle down as fast as they think it does; and that they are consistently underestimating tail risk.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.

User avatar
bobcat2
Posts: 5337
Joined: Tue Feb 20, 2007 3:27 pm
Location: just barely Outside the Beltway

Re: Misbehaviour of markets and the Black Swan

Post by bobcat2 » Tue Jul 02, 2019 9:00 am

I don't think this is very important, but given others apparently do, I will gingerly weigh in on the discussion of Fama's doctoral advisors. It appears to me that both sides are partially correct. Fama's formal doctoral advisors were Merton Miller and Harry Roberts. But because Fama's doctoral thesis was an emperical study showing that stock returns were fat tailed - Mandelbrot, who was a frequent visitor at the U of Chicago at that time, was an important informal advisor on Fama's doctoral thesis.

You can read more about this
here - https://en.wikipedia.org/wiki/Eugene_Fama#cite_note-6
[Fama's] M.B.A. and Ph.D. came from the Booth School of Business at the University of Chicago in economics and finance. His doctoral supervisors were Nobel prize winner Merton Miller and Harry Roberts, but Benoit Mandelbrot was also an important influence.

and here - https://famafrench.dimensional.com/essa ... nance.aspx
During my second year at Chicago, with an end to course work and prelims in sight, I started to attend the Econometrics Workshop, at that time the hotbed for research in finance. Merton Miller had recently joined the Chicago faculty and was a regular participant, along with Harry Roberts and Lester Telser. Benoit Mandelbrot was an occasional visitor. Benoit presented in the workshop several times, and in leisurely strolls around campus, I learned lots from him about fat-tailed stable distributions and their apparent relevance in a wide range of economic and physical phenomena. Merton Miller became my mentor in finance and economics (and remained so throughout his lifetime). Harry Roberts, a statistician, instilled a philosophy for empirical work that has been my north star throughout my career.

Miller, Roberts, Telser, and Mandelbrot were intensely involved in the burgeoning work on the behavior of stock prices (facilitated by the arrival of the first reasonably powerful computers). At the end of my second year at Chicago, it came time to write a thesis, and I went to Miller with five topics. Mert always had uncanny insight about research ideas likely to succeed. He gently stomped on four of my topics, but was excited by the fifth. From my work for Harry Ernst at Tufts, I had daily data on the 30 Dow-Jones Industrial Stocks. I proposed to produce detailed evidence on (1) Mandelbrot's hypothesis that stock returns conform to non-normal (fat-tailed) stable distributions and (2) the time-series properties of returns. There was existing work on both topics, but I promised a unifying perspective and a leap in the range of data brought to bear.

Vindicating Mandelbrot, my thesis (Fama 1965a) shows (in nauseating detail) that distributions of stock returns are fat-tailed: there are far more outliers than would be expected from normal distributions - a fact reconfirmed in subsequent market episodes, including the most recent. Given the accusations of ignorance on this score recently thrown our way in the popular media, it is worth emphasizing that academics in finance have been aware of the fat tails phenomenon in asset returns for about 50 years.
BobK
In finance risk is defined as uncertainty that is consequential (nontrivial). | The two main methods of dealing with financial risk are the matching of assets to goals & diversifying.

User avatar
bobcat2
Posts: 5337
Joined: Tue Feb 20, 2007 3:27 pm
Location: just barely Outside the Beltway

Re: Misbehaviour of markets and the Black Swan

Post by bobcat2 » Tue Jul 02, 2019 9:24 am

grabiner wrote:Note that this is a distribution of daily returns. With relatively low correlation between daily returns, the returns over longer periods are closer to a normal distribution.

This is a general principle in statistics. The Central Limit Theorem says that under certain conditions (independent identical distributions, for example), you can prove that the sum of a large number of events will have a nearly-normal distribution. But even when the conditions do not hold, you expect a normal distribution in practice when you add many small random variables to get a large one, such as 250 daily stock returns to get an annual return.
Annual returns are typically fairly well represented by the log-normal distribution and exhibit little volatility clustering. There are two principal reasons that annual returns are better represented by the log-normal distribution rather than the normal distribution. Namely, the compounding of the returns and the zero lower bound on stock returns.

Annual returns are slightly more fat tailed than a log-normal distribution but that can be handled by applying a GARCH(1,1) model to the log-normal or simply cheating and adding a couple of percent to the standard deviation. Sometimes followers of Mandelbrot claim that the Central Limit Theorem doesn't hold in the case of annual stock returns because the returns are not independent of one another. But high frequency returns don't have to be strictly independent for the annual returns to be close to log-normal. So whether the annual returns are approximately log-normal becomes an empirical question. On the grounds of observation and experience, aka empirically, Mandelbrot and his followers appear to be wrong. When you analyze the historical annual US stock market data it appears to be only slightly more fat tailed than a log-normal distribution.

BobK
In finance risk is defined as uncertainty that is consequential (nontrivial). | The two main methods of dealing with financial risk are the matching of assets to goals & diversifying.

protagonist
Posts: 5829
Joined: Sun Dec 26, 2010 12:47 pm

Re: Misbehaviour of markets and the Black Swan

Post by protagonist » Tue Jul 02, 2019 9:47 am

nisiprius wrote:
Sun Jun 30, 2019 5:37 pm
Yes. The Mandelbrot book in particular amazed me. I felt as if the scales had fallen from my eyes. I think it is interesting how rarely mainstream investment writers and experts talk about it. "Nihilistic" is exactly the word I would use, because if Mandelbrot is right, 99% of the financial economics used in investing is nonsense. Furthermore, I don't think the mainstream has any answer to it. Their response is just to look away and pretend it isn't there.

The word "obscene" is literally derived from words meaning "look away" or "avert your eyes," something so repellent you can't bear to look at it. To the investing industry, The Misbehavior of Markets is an obscene book.

Fooled by Randomness and The Black Swan were discussed quite a lot in this forum when they came out, and "black swan" has become part of the language. I'm about halfway through Taleb's book, Skin in the Game. I have mixed feelings. I feel that his writing does drift off into self-indulgence and unbridled ego. I find his books stimulating to read and full of interesting ideas, but I don't put him in the same class as Mandelbrot.
+1. I agree strongly with all of the above. Read and re-read Nisiprius' post because, imho, he hit the nail on the head.

I view Taleb (ancecdotally from what people have written about his work) as essentially popularizing chaos and complexity theory for the masses (with some nonsense thrown in)...nothing really new there- at least not enough to compel me to buy his books. That said he seems to be popularizing a very important message that is lost on many.

To understand the markets one must understand the dynamics of complex,nonlinear systems....rather than trying to predict the future based on inadequate prior data , and no testable and falsifiable hypotheses.

This article is an excellent primer: https://static.squarespace.com/static/5 ... 75/cce.pdf

For those interested in exploring further, the folks at Santa Fe Institute (pioneers in chaos and complexity research) have a number of rotating online courses (many free) that are a good springboard. I don't know if any specifically focus on economics but there are introductory courses to the field of complexity science. https://www.complexityexplorer.org/news ... e-schedule "The Santa Fe Institute is leading the world in complexity science, with a mixed group of physicists, biologists, economists, political scientists, computer experts, and mathematicians working together." (-Freeman Dyson in The New York Review of Books).
Last edited by protagonist on Tue Jul 02, 2019 10:00 am, edited 1 time in total.

latesaver
Posts: 158
Joined: Thu Aug 03, 2017 3:35 pm

Re: Misbehaviour of markets and the Black Swan

Post by latesaver » Tue Jul 02, 2019 12:39 pm

alex_686 wrote:
Sun Jun 30, 2019 7:00 pm
So, let me modestly take the other side.

When it comes to standard deviations, I am a "glass half full" type of guy - or to be precise, 95% full. It provides a robust framework, independent of asset time and time horizon. It is simple and robust. And it is good out to 2 to 3 standard deviations. And the defects and shortcomings are well known.

I like Mandelbrot. However it adds lots of complexity and sophistication for very little gain. Plus you loss a lot of universality.

I dislike Nassim. His criticism are valid but were widely known. The problem is that he does not offer much in the way of solutions.
x2, +1, etc

Taleb spends too much time whining about not winning the Nobel prize while Thaler, Kahneman, et al did/do.

Black swans are interesting but i don't see how to utilize them in a sensible fashion into a BH style portfolio. my recollection from the book was to take little risk with 80-90% of the portfolio and barbell it with very high risk investments for the balance. Not interested.

User avatar
nisiprius
Advisory Board
Posts: 38471
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: Misbehaviour of markets and the Black Swan

Post by nisiprius » Tue Jul 02, 2019 2:25 pm

latesaver wrote:
Tue Jul 02, 2019 12:39 pm
...Black swans are interesting but i don't see how to utilize them in a sensible fashion into a BH style portfolio. my recollection from the book was to take little risk with 80-90% of the portfolio and barbell it with very high risk investments for the balance...
I remember, near the end of the book, when he'd been saying something about "positive black swan," I said to myself "but this is all just a fancy way of saying 'I like to play long shots.'"

My dislike for Taleb spiked when I made the mistake of reading The Bed of Procrustes: Philosophical and Practical Aphorisms. I wrote a parody review of it in this forum, which didn't get me nearly as many "attaboys!" as I hoped for. I still keep linking to it in hopes someone will think it's funny: The Garden of Persephone: Lophosaurical and Whimsical Pittosporums, by Aeneas Massif Table. I like to think that the problem with it is that it is not funny unless you've actually read The Bed of Procrustes, which isn't worth reading.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.

bgf
Posts: 959
Joined: Fri Nov 10, 2017 9:35 am

Re: Misbehaviour of markets and the Black Swan

Post by bgf » Tue Jul 02, 2019 2:57 pm

nisiprius wrote:
Tue Jul 02, 2019 2:25 pm
latesaver wrote:
Tue Jul 02, 2019 12:39 pm
...Black swans are interesting but i don't see how to utilize them in a sensible fashion into a BH style portfolio. my recollection from the book was to take little risk with 80-90% of the portfolio and barbell it with very high risk investments for the balance...
I remember, near the end of the book, when he'd been saying something about "positive black swan," I said to myself "but this is all just a fancy way of saying 'I like to play long shots.'"

My dislike for Taleb spiked when I made the mistake of reading The Bed of Procrustes: Philosophical and Practical Aphorisms. I wrote a parody review of it in this forum, which didn't get me nearly as many "attaboys!" as I hoped for. I still keep linking to it in hopes someone will think it's funny: The Garden of Persephone: Lophosaurical and Whimsical Pittosporums, by Aeneas Massif Table. I like to think that the problem with it is that it is not funny unless you've actually read The Bed of Procrustes, which isn't worth reading.
the memorable parts are the not-subtle burns -

“There is no clearer sign of failure than a middle-aged man boasting of his performance in college.”

“If you know, in the morning, what your day looks like with any precision, you are a little bit dead - the more precision, the more dead you are.”

“The best test of whether someone is extremely stupid (or extremely wise) is whether financial and political news makes sense to him.”

“Never take investment advice from someone who has to work for a living.”

what is interesting, is that most of the stuff he says reveals more about him, and his insecurities, than others. which you might be able to tell from the above quotes.
“TE OCCIDERE POSSUNT SED TE EDERE NON POSSUNT NEFAS EST"

Angst
Posts: 1997
Joined: Sat Jun 09, 2007 11:31 am

Re: Misbehaviour of markets and the Black Swan

Post by Angst » Wed Jul 03, 2019 10:29 am

wrote:
Tue Jul 02, 2019 8:37 am
bgf wrote:
Tue Jul 02, 2019 7:42 am
mandelbrot, in his book, states in no uncertain terms that he was Fama's thesis adviser.
Given what I know about his personality, this claim doesn't surprise me.
I don't attach much importance to points simply made with a slur.
bobcat2 wrote:
Tue Jul 02, 2019 9:00 am
I don't think this is very important, but given others apparently do, I will gingerly weigh in on the discussion of Fama's doctoral advisors.

[Snip...]
Thank you BobK (and Nisi, OP, et al) for your thoughtful and patient participation in this thread and for trying to keep the focus on the science. As in the past, it's been interesting. I believe it's worthwhile revisiting the discussion of Mandelbrot's ideas every now and then. And there are always newbies who have no idea.

User avatar
Topic Author
305pelusa
Posts: 289
Joined: Fri Nov 16, 2018 10:20 pm

Re: Misbehaviour of markets and the Black Swan

Post by 305pelusa » Wed Jul 03, 2019 4:48 pm

Mkay so as far as actionable investment strategies to take into account the information of these books, I'm seeing:

1) Stick to market index funds. If I'm understanding the messages of these books correctly, I'm not too convinced this would be a strategy that takes the material into consideration.

2) Stick to Larry's investment advice. Since he has read the books, I presume his advice takes the material into account. The part I find inconsistent here is that alternative investments seem like the exact opposite of what Nassim would recommend. These funds expose you to left tail risk (negative Black Swans) instead of right tail returns (positive Black Swans). I haven't read Larry's Black Swan book though. Presumably, there he'd explain just how unlikely it is for all of the left tails to happen at the same time while you'd be protected from any one Black Swan due to the diversification.

To be absolutely clear, I am not saying the above are bad choices. But I'm not sure these really take into account an investment mindset if you believe the material in the books.

I saw the advice of 90% T-Bills, 10% of aggressive private equity or maybe VC. This makes some sense to me although it's almost diametrically opposite from the above.

I guess I was looking for something somewhere in the middle. Not necessarily a full-on Nassim portfolio, but perhaps small modifications to apply the material to an extent. I'm all ears if anyone has gone on to do this.

protagonist
Posts: 5829
Joined: Sun Dec 26, 2010 12:47 pm

Re: Misbehaviour of markets and the Black Swan

Post by protagonist » Wed Jul 03, 2019 4:58 pm

nisiprius wrote:
Tue Jul 02, 2019 2:25 pm
latesaver wrote:
Tue Jul 02, 2019 12:39 pm
...Black swans are interesting but i don't see how to utilize them in a sensible fashion into a BH style portfolio. my recollection from the book was to take little risk with 80-90% of the portfolio and barbell it with very high risk investments for the balance...
I remember, near the end of the book, when he'd been saying something about "positive black swan," I said to myself "but this is all just a fancy way of saying 'I like to play long shots.'"

My dislike for Taleb spiked when I made the mistake of reading The Bed of Procrustes: Philosophical and Practical Aphorisms. I wrote a parody review of it in this forum, which didn't get me nearly as many "attaboys!" as I hoped for. I still keep linking to it in hopes someone will think it's funny: The Garden of Persephone: Lophosaurical and Whimsical Pittosporums, by Aeneas Massif Table. I like to think that the problem with it is that it is not funny unless you've actually read The Bed of Procrustes, which isn't worth reading.
I don't like him either, but if he really said "In springtime, seed; in summer, cultivate; in fall, windsurf.", then my opinion of him just went up a notch.

On the other hand, if you wrote it, then my opinion of you did. (note: I had a good opinion of you to start with...much better than my opinion of Taleb.)
Last edited by protagonist on Wed Jul 03, 2019 5:41 pm, edited 1 time in total.

protagonist
Posts: 5829
Joined: Sun Dec 26, 2010 12:47 pm

Re: Misbehaviour of markets and the Black Swan

Post by protagonist » Wed Jul 03, 2019 5:37 pm

Contrary to other's opinions, I do find chaos/complexity (eg: Mandelbrot) actionable.

1. There is nothing wrong with living in a house of cards if they are the only houses available. But knowing what it is made of keeps one from buying into the "financial porn", and keeps one from taking baseless statistics and the latest untestable hypotheses seriously when making decisions. The pig in the house of straw who recognizes his home's vulnerability might be able to escape the wind better than the one who thinks that it is secure.

2. It is, in a way, liberating to realize that you DON'T have control, so that if your great system ultimately fails you, you don't blame it on your own stupidity. To quote the Incredible String Band paraphrasing Shakespeare. "All the world is but a play. Be thou the joyful player." And to quote Shakespeare directly, "There are more things in heaven and Earth, Horatio, / Than are dreamt of in your philosophy". Or as Robert Burns said, "The best laid schemes o' mice an' men / Gang aft a-gley.” Admitting that to one's self would at the very least reduce the suicide rate. So no, we have no reason to buy into what range of net worth we might have 30 or 50 years from now based on some Monte Carlo studies relying on less than a century of past performance, or to think that a market that crashes will bounce back like it did in 2008. Admitting lack of control makes us more resilient to the vicissitudes, which reduces the inclination to panic when things go aglay.

Post Reply