Nobel prize for fat tail theory

Financial Independence/Retire Early -- Learn How!
Post Reply
Kramer
* Rookie
Posts: 18
Joined: Mon Aug 25, 2003 3:53 pm

Nobel prize for fat tail theory

Post by Kramer »

One of the recent discussions was about leptokurtic (yep, it was the correct word) or fat-tail distributions of stock market returns. Combined with the tendency of the stock market to exhibit return to mean behavior over long periods of time. This makes it difficult to use monte carlo techniques for SWR analysis, for instance, because stock returns cannot be forecast with conventional Normal (Gaussian) distributions.

Turns out that a couple of really smart economists just won the nobel prize for figuring this problem out.

Two economists (Engle, Granger) have been awarded the 2003 Nobel prize in economics for their pioneering work in developing statistical methods that make economic models more realistic and useful.

Engle's insight was to notice that the volatility of many financial or economic variables changes over time -- periods of great change are followed by calmer periods -- whereas earlier statistical models had assumed a constant rate of volatility.


See section 3 of the Nobel org's paper synopsis:

http://www.nobel.se/economics/laureates/2003/ecoadv.pdf

http://socialize.morningstar.com/NewSoc ... 1065740940

http://cbs.marketwatch.com/news/story.a ... eid=intuit


If someone wanted to condense and summarize this for the rest of us, or enhance their simulations, that would be great :D

Kramer
caseynshan
* Rookie
Posts: 16
Joined: Fri May 30, 2003 3:00 am
Location: Denver

Post by caseynshan »

I don't know why, but i found this sentence amusing....

"Engle's breakthrough concept, known as autoregressive conditional heteroskedasticity, is now an accepted part of every statisticians' tool kit"
therealchips
*** Veteran
Posts: 174
Joined: Sat Jan 04, 2003 4:00 am
Location: Henderson, Nevada, USA

A straightforward buy-and-hold kind of an investor

Post by therealchips »

I liked the disconnect between these two:
"It affords you a way to statistically and formally summarize and characterize volatility over time. The markets are very volatile, and up until Engle's model, they (researchers) knew that fact but not how to deal with it in a formal way," said Tim Bollerslev, a Duke economics professor. "It really kind of radically changed the way Wall Street people think about risks."

Engle added that he does not use his model in his personal investing, calling himself a "straightforward buy-and-hold kind of an investor."

USAToday links seem to disappear quickly, but here it is:
http://www.usatoday.com/money/economy/2 ... obel_x.htm
He who has lived obscurely and quietly has lived well. [Latin: Bene qui latuit, bene vixit.]

Chips
therealchips
*** Veteran
Posts: 174
Joined: Sat Jan 04, 2003 4:00 am
Location: Henderson, Nevada, USA

Influence of his statistical analysis on investment strategy

Post by therealchips »

Maybe this link will last longer:
http://www.nynewsday.com/news/local/man ... -manhattan
Basically, he developed a way to dissect statistics by measuring factors that influenced them, then looking for relationships among those factors. "You start with two statistics: the number of ducks, which is growing, and the gross domestic product, which is also growing," explained Dean Baker, an economist at the Center for Economic and Policy Research in Washington.

"But before you conclude that one leads to the other, Engle developed the statistical analysis to help determine the factors that led to each trend and determine from that whether there was any relationship," Baker said. "It was very important work."

The tools have been applied to financial markets and helped investors price stocks, derivatives, stock options and weigh risk versus return.

Has the exotic statistical analysis influenced his investment strategy? No, replied Engle, who splits the $1.3-million prize with Granger. "I'm really a pretty straight-forward, buy-and-hold kind of investor."
He who has lived obscurely and quietly has lived well. [Latin: Bene qui latuit, bene vixit.]

Chips
Kramer
* Rookie
Posts: 18
Joined: Mon Aug 25, 2003 3:53 pm

Post by Kramer »

Yep, this does not affect my buy and hold strategy during either accumulation or during retirement. But it is central to long term survivability of diversified portfolios, and how much you can withdraw out of them over time, and what assets they should consist of. These economists don't have to worry about it because they have got cushy university pensions, not to mention about a million more dollars for winning :wink:

I was on William Sharpe's web site ( http://www.stanford.edu/~wfsharpe/home.htm ), and he had a pointer to the GARCH toolbox for MATLAB (an engineering simulation tool), a special toolkit available for purchase for MATLAB that enables one to do sophisticated monte carlo withdrawal scenarios, among other things. I have MATLAB on my PC and am a skilled user. Unfortunately, I won't have time to pursue this for awhile, as I am taking a pretty tough course on the side right now, at the same school, to prepare for what might be inevitable layoff from my Silicon Valley job. :?

Kramer
wanderer
*** Veteran
Posts: 363
Joined: Tue Nov 26, 2002 9:33 am
Location: anytown, usa

Post by wanderer »

These economists don't have to worry about it because they have got cushy university pensions,

People actually care about what these chaps profess. And folks like Shiller and others. there is a second, third and fourth tier of academician, however.

The average PhD doesn't have enough sense to pound sand, IMO. Most of my peers are unpublishable in the reputable journals. So they play mickey mouse games where they publish in each other's non descript journals (ones on whose peer review committee they sit). Then their articles go to the place where ideas that never should have wasted pulp go to die, unreferenced and unlamented.

They have a lot to worry about because this is their best job, they were essentially unemployed for ten years, with little to invest, and armed with little common sense, during the greatest stock run up in history. And they walked away with tons of student loan debt.

Forced to go back to the States, or their country of origin, they would be relatively unemployable. Most of the English PhDs would be waiters.

not to mention about a million more dollars for winning

$1.3mmUS, split two ways, doesn't go as far as it used to. :wink:
regards,

wanderer

The field has eyes / the wood has ears / I will see / be silent and hear
WiseNLucky
** Regular
Posts: 84
Joined: Tue Nov 26, 2002 3:59 am
Location: Florida

Post by WiseNLucky »

Nobel prize for fat tail . . .


There's hope for me yet!
WiseNLucky

I just wish everyone could step back and get less car and less house then they want, and realize they don't NEED more. -- NeuroFool
therealchips
*** Veteran
Posts: 174
Joined: Sat Jan 04, 2003 4:00 am
Location: Henderson, Nevada, USA

Who is fat in the tail? Who is thick-skinned?

Post by therealchips »

Kramer:
One of the recent discussions was about leptokurtic (yep, it was the correct word) or fat-tail distributions of stock market returns.


http://arbl.cvmbs.colostate.edu/hbooks/ ... eptin.html
Leptin (from the Greek leptos, meaning thin) is a protein hormone with important effects in regulating body weight, metabolism and reproductive function.


The Greek for fat is "pachy" as in pachyderms, the thick-skinned ones. How do fat-tailed distributions get a word for "thin" in their designation? It seems that the thinness of leptokurtic distritributions refers to their being thinner in their central regions, and having a higher peak around the mean, compared to a normal or Gaussian distribution with the same mean and variance. [I will insert here a graph of a leptokurtic distribution superimposed on a Gaussian distribution with the same mean and variance, whenever I find one online.]
He who has lived obscurely and quietly has lived well. [Latin: Bene qui latuit, bene vixit.]

Chips
therealchips
*** Veteran
Posts: 174
Joined: Sat Jan 04, 2003 4:00 am
Location: Henderson, Nevada, USA

Post by therealchips »

$1.3mmUS, split two ways, doesn't go as far as it used to, and Nobel prizes became taxable income for US citizens in a recent "reform". The split will be three ways.
He who has lived obscurely and quietly has lived well. [Latin: Bene qui latuit, bene vixit.]

Chips
User avatar
ataloss
**** Heavy Hitter
Posts: 559
Joined: Mon Nov 25, 2002 3:00 am

Post by ataloss »

Epsteins in barrons
So much of what passes for economic news and analysis consists of watershed-of-the-week reportage whereby some line graph of recent vintage is either hailed or bewailed as yet another sign of historic change. A systematic look at the pattern of previous decades usually reveals the been-there, done-that nature of the new find.

The high-powered magnifiers built, so to speak, by econometricians Engle and Granger have helped sharpen the focus through which past patterns can be observed. (Econometrics: the statistical analysis of economic variables; variable: a number subject to change.)

Engle is credited with developing methods of analyzing cases in which the volatility of, say, stock prices rises or falls over time. Among other things, this improves the accuracy of stock-options valuation.

Granger has enhanced the ability to determine whether the correlation between two variables implies that one causes the other, or whether the variables are correlated simply because they happen to be traveling in the same direction.

Great stuff; but now let's add perspective. What Granger and Engle do is ultimately art, not science -- except when it makes use of laws of economics that actually are scientific: for example, the law of supply and demand being on a par with, well, the law of gravity.
Have fun.

Ataloss
Post Reply