Fooled by Randomness
- Thread starter oddiduro
- Start date
Taleb's thesis is that humans - regardless of education and training - have a difficult time coming to terms with the concepts of randomness.
It is a decent book to read, if it makes you question results and data from a different perspective.
Because I read it a while back, I cannot remember what his take on TA was, but I do recall him making a few points about traders optimizing the crap out of their systems in Tradestation (no less), and some of the "high-priests" of finance get a hammering as well for choosing to "model" the real world with mythical statistical distributions
amongst other things. if anything, Prof Merton was mentioned quite a few times (which is a bit stiff - since Bob is actually a really nice guy - you wouldn't know he had won the Nobel prize by having a chat with him.. but I digress)
{Edit...}
Kinda liked his take on expectation, and he seemed to have a thing about Dentists, and dental "pangs"...
A bit more here...
http://www.wilmott.com/detail.cfm?articleID=208
It is a decent book to read, if it makes you question results and data from a different perspective.
Because I read it a while back, I cannot remember what his take on TA was, but I do recall him making a few points about traders optimizing the crap out of their systems in Tradestation (no less), and some of the "high-priests" of finance get a hammering as well for choosing to "model" the real world with mythical statistical distributions
amongst other things. if anything, Prof Merton was mentioned quite a few times (which is a bit stiff - since Bob is actually a really nice guy - you wouldn't know he had won the Nobel prize by having a chat with him.. but I digress){Edit...}
Kinda liked his take on expectation, and he seemed to have a thing about Dentists, and dental "pangs"...
A bit more here...
http://www.wilmott.com/detail.cfm?articleID=208
Quote from Equalizer:
Taleb's thesis is that humans - regardless of education and training - have a difficult time coming to terms with the concepts of randomness.
It is a decent book to read, if it makes you question results and data from a different perspective.
Because I read it a while back, I cannot remember what his take on TA was, but I do recall him making a few points about traders optimizing the crap out of their systems in Tradestation (no less), and some of the "high-priests" of finance get a hammering as well for choosing to "model" the real world with mythical statistical distributions
Well, LTCM did almost end the financial world as we know it
If Talib is right, basically any kind of hindsight analysis is pretty much pointless.
i disagree.
IMHO, taleb's point is that in a world where we don't really know what the distribution of returns looks like then there is a structural imbalance in the market towards "selling volatility", trading "mean-reversions" or other kinds of programs which take advantage of "normal" market conditions.
he suggests that the cumulative small profits earned by this kind of trading do not cover the risk of an eventual "black swan" (ie very very rare, out-of-paradigm) event.
i think of it this way - in a world where nothing bad is happening or expected to happen or where some external agent is deemed to guarantee bail-out, insurance seems expensive, so less people buy it or if the demand stays constant then the supply of insurers will increase to take advantage of the free-money on offer. either way the insurance premiums will fall and not cover the "real" unknown risk that exists.
his point is that his kind of trading: long-vega/gamma/big trend-seeking or more generally trading with a pre-defined stop which suffers lots of small losses (to the other side above) eventually gets the big payoff. he's very respectful of street-smart types (he did trade in the pits in chicago) but very anti semi-quant types who expound and use engineering/mathematical methods without really understanding their bases.
Q1
IMHO, taleb's point is that in a world where we don't really know what the distribution of returns looks like then there is a structural imbalance in the market towards "selling volatility", trading "mean-reversions" or other kinds of programs which take advantage of "normal" market conditions.
he suggests that the cumulative small profits earned by this kind of trading do not cover the risk of an eventual "black swan" (ie very very rare, out-of-paradigm) event.
i think of it this way - in a world where nothing bad is happening or expected to happen or where some external agent is deemed to guarantee bail-out, insurance seems expensive, so less people buy it or if the demand stays constant then the supply of insurers will increase to take advantage of the free-money on offer. either way the insurance premiums will fall and not cover the "real" unknown risk that exists.
his point is that his kind of trading: long-vega/gamma/big trend-seeking or more generally trading with a pre-defined stop which suffers lots of small losses (to the other side above) eventually gets the big payoff. he's very respectful of street-smart types (he did trade in the pits in chicago) but very anti semi-quant types who expound and use engineering/mathematical methods without really understanding their bases.
Q1
Quote from Quiet1:
..............
his point is that his kind of trading: long-vega/gamma/big trend-seeking or more generally trading with a pre-defined stop which suffers lots of small losses (to the other side above) eventually gets the big payoff. he's very respectful of street-smart types (he did trade in the pits in chicago) but very anti semi-quant types who expound and use engineering/mathematical methods without really understanding their bases.
Q1
i wonder what he would make of the current trading dynamics occurring right now. basically the NYSE is over run by trading bots(the quants) and the street smart guys are at a huge disadvantage. will this dynamic correct itself??? if so, what would it take and how does that scenario play itself out?
Quote from Quiet1:
things may never get better for human traders. why should they? the current set of market agents will continue to evolve either through gradual incremental change or externally induced shock.
twas always thus...
:eek:
Q1
actually i would have agreed with this statement a year or two ago but i believe we are currently at the bottom for human traders. there are several reasons i say this. firstly, my own experience...i.e. my p&l has leveled off for quite sometime. also, i have reason to believe the big bot players are being squeezed by each other now and that their cost are rock bottom and are not going anywhere, thus what you see is what you get. i believe it is either goldman or morgan that is starting to lay off traders...not exactly a sign of their future growth hopes. in this devolution or atrophy i have become extremely cost efficient as i am sure many other "human" traders have also. although the big firms stated commission is less than mine, at the end of the day my cost per share probably is competitive. here's my tie in to taleb, we have yet to hear of a black swan event in regards to these big bot firms, if/when it occurs what will be the fall out? will the current quant business model continue to be employed? will there be new laws/regulations that make the current business model unprofitable? regardless, i refer back to my earlier statement, my p&l has leveled off for quite sometime...... i seriously doubt these bots can do anymore damage to my profitability. any future changes i can see coming will not hurt me but will probably affect them.
Very good summary of the "Taleb Thesis".
I think it's a valid one.
I think it's a valid one.
Quote from Quiet1:
i disagree.
IMHO, taleb's point is that in a world where we don't really know what the distribution of returns looks like then there is a structural imbalance in the market towards "selling volatility", trading "mean-reversions" or other kinds of programs which take advantage of "normal" market conditions.
he suggests that the cumulative small profits earned by this kind of trading do not cover the risk of an eventual "black swan" (ie very very rare, out-of-paradigm) event.
i think of it this way - in a world where nothing bad is happening or expected to happen or where some external agent is deemed to guarantee bail-out, insurance seems expensive, so less people buy it or if the demand stays constant then the supply of insurers will increase to take advantage of the free-money on offer. either way the insurance premiums will fall and not cover the "real" unknown risk that exists.
his point is that his kind of trading: long-vega/gamma/big trend-seeking or more generally trading with a pre-defined stop which suffers lots of small losses (to the other side above) eventually gets the big payoff. he's very respectful of street-smart types (he did trade in the pits in chicago) but very anti semi-quant types who expound and use engineering/mathematical methods without really understanding their bases.
Q1