Serving the Quantitative Finance Community

 
User avatar
MobPsycho
Topic Author
Posts: 1946
Joined: March 20th, 2002, 2:53 pm

.

October 1st, 2002, 2:39 pm

Last edited by MobPsycho on August 17th, 2003, 10:00 pm, edited 1 time in total.
 
User avatar
akimon
Posts: 259
Joined: May 28th, 2002, 2:38 pm

.

October 1st, 2002, 5:38 pm

I'm confused about your points. So one "gets lucky" if one profits from a game that has an expect earnings of less than the entry price. Someone can make money and get lucky at the same time.But are you saying that "making money" is the act of participating in a game that has expected returns in ones favour (professional poker), or does "making money" involve increasing the total wealth of the world (writing the program)?
 
User avatar
MobPsycho
Topic Author
Posts: 1946
Joined: March 20th, 2002, 2:53 pm

.

October 1st, 2002, 6:16 pm

Last edited by MobPsycho on August 17th, 2003, 10:00 pm, edited 1 time in total.
 
User avatar
Traden4Alpha
Posts: 23951
Joined: September 20th, 2002, 8:30 pm

.

October 3rd, 2002, 9:10 pm

RE; lucky vs. smart: Isn't the problem that the distinction between lucky vs. smart is impossible to resolve based on a single data point? That one person in a million creates a software company with a billion dollar market cap or one trader in a million reaps a zillion% return are both indistinguishable from statistical flukes. But, if that same person creates two "flukes" in a row or 3 or an unending stream, the probability of "luck" drops and the probability of "smart" increases. Admittedly, this all must be compared to the chance of being lucky n-times in a row under the assumption of random, independent trials. And, I won't touch the issue that the world is not a series of random, independent trials (that being lucky once creates success that feeds on itself) .RE: the ludicrous definition of average: I agree about the tautological nature of the definition of above-average. This is made doubly bad by the selectivity of participation. If a group of people choose to be in the market when it has a high average return, and they all earn that market average for that year, are they lucky, smart, or average WRT those that did not put money in the market? I ask, because by this definition, I am none too smart as I did not invest in internet companies when it was "smart" to do so. Finally, are the people who are currently moving their depleted portfolios over into over-priced real-estate an example of "bad luck" or market-average returns? RE: Robinson Crusoe: You could also view this as an example of survivorship bias -- the book could not exist if the castaway been left on an uninhabitable island.RE: efficient markets: You are totally right about the phenomenon of copying the pioneers - the pioneer gets lucky only to see their profits dwindle under the onslaught of the bandwagon. But, with the markets, are there forms of sustainable profiteering based on counterintuitive strategies? Could some participants earn excess profits in perpetuity because others refuse to follow their strange methods? Also, maybe the real examples of smart are those that consistently stay one step ahead of the imitators.I'm feeling lucky to have run into a forum populated by smart people while I'm Traden4Alpha
 
User avatar
JabairuStork
Posts: 947
Joined: February 27th, 2002, 12:45 pm

.

October 4th, 2002, 12:17 pm

traden4alpha,I agree with you (and Bertrand Russell) that inference is one of the most powerful means we have for understanding our universe. As one person realizes more successes, the probability that he is good (vs. lucky) increases. However, it is important to keep in mind that a guy who makes $1 billion starting from nothing exists in a different probability space from the guy who makes his second billion. The second guy should be considered in a reduced probability space of billionaires who have made another billion.It's a good question about whether any one person or firm (even if we assume that they are good because they are skilled and luck has nothing to do with it) can consistently and perpetually earn above market returns. Theory would suggest that this should be impossible, although (limited) observation suggests otherwise. I guess it raises questions of scale invariance - should the same principles that govern high frequency realizations govern lower frequencies?
 
User avatar
Johnny
Posts: 2431
Joined: October 18th, 2001, 3:26 pm

.

October 4th, 2002, 12:23 pm

"Theory would suggest that this should be impossible"Which theory?
 
User avatar
Traden4Alpha
Posts: 23951
Joined: September 20th, 2002, 8:30 pm

.

October 4th, 2002, 1:41 pm

JabairuStork,I totally agree with you that the second billion is made within a different probability space. It is the causal factors that underpin that different probability space that concern me. Success (whether by luck or by smarts) seems self-amplifying.The problem is that the average bounded-rational evaluator is incapable of distinguishing between lucky and smart forms of first-time success (and usually errs toward the presumption of intelligence). Thus, once someone is successful, they gain access to resources unavailable to the not-successful. This is most obvious for the case of serial entrepreneurs -- being successful in one business gives them an advantage in attracting both capital and skilled personnel to follow-on businesses. But even traders enjoy a form of success to the successful. Successful traders have the asset base needed for proper risk management and value-added services. I'm appalled by the number of low-cap traders who want to learn trading use $5000 for fully leveraged trading of the e-Minis or who swing-trade a single stock because they cannot afford to diversify. In contrast, better capitalized traders can diversify. Moreover, successful traders also can access premium services that give them further advantages (e.g., high-end data services, dedicated high-speed connections to markets, lower-cost brokerage services, priority customer service, lower margin interest rates, etc.). Another case of success to the successful.And, at the upper-end are world-renowned investors/traders (e.g., Buffet, Soros, etc.) whose actions create favorable self-fulfilling prophecies. Although I would never wish to detract from their success, I would argue that they could use a dart-board to pick take-over targets and still earn market-beating returns.RE: consistently and perpetually earn above market returns: I think this becomes a matter of deciding which of four hypothesized failure mechanisms is the cause of inconsistent returns. H1) Failure is due to the end of luck: If the markets are random, then success and failure is a simple stochastic process. Given the large numbers of participants in the markets, it is inevitable that some will appear to produce above average returns for above average lengths of time. Yet, at each time-point they would have the same chance of future success as everyone else.H2) Failure is due to rigidity: If the markets contain slowly-varying non-stationary patterns, then above average returns are possible by those that synchronize with those patterns. Yet with increasing success would come increasing confidence. Increasing confidence would lead to increasing risk-taking (e.g., use of margin) and increasing rigidity. An eventual shift in the market's underlying patterns would lead to above average losses as the previously successful trader stubbornly adheres to previously successful methods.H2) Failure is due to excessive size: The market may contain consistently exploitable patterns of bounded size. As the successful trader compounds their assets (by trading or by soliciting other peoples money) they would eventually exceed the profitable capacity of the pattern. They would appear to lose their edge and slip back into mediocrity.H3) Failure is due to the rise of imitators: As successful traders become more visible, they become the subjects of imitators. This failure mechanism is exacerbated if the trader operates within an organizational context (e.g. proprietary trading firm or professional investment/trading firm). Peers and coworkers will inevitably learn the successful trader's techniques and imitate them. Personal ego also leads to the rise of imitators as the pride of the successful leads them to grant interviews and pontificate on their success. RE: Scale invariance: Interesting issue! It would seem to me that the high-frequency realizations occur within some approximately stationary stochastic environment, while the lower-frequency realizations are governed by the more deterministic self-modifying/adaptive nature of the markets. Admittedly, that statement is more descriptive, than prescriptive. The challenge for the dedicated trader is to uncover and exploit the structure of the temporarily-stationary stochastic environment while adapting to the inevitable shifts in the market's pattern of price action.Thanks for making me exert my bounded rationality on this problem. I'm trying to determine if H1 is true, while avoiding H2,H3, and H4.
Last edited by Traden4Alpha on October 3rd, 2002, 10:00 pm, edited 1 time in total.
 
User avatar
Aaron
Posts: 6433
Joined: July 23rd, 2001, 3:46 pm

.

October 4th, 2002, 4:58 pm

To paraphrase Herodotus quoting Solon: "call no man lucky until he's broke." We do not know what the future has in store, and recent years have seen hundred-billion-dollar piles of wealth evaporate overnight. While this scale of disaster is recent, history is full of smart and lucky people who make gigantic fortunes and then go bankrupt, often many times. In fact, it's a job requirement if you want to be a real estate developer.
 
User avatar
Traden4Alpha
Posts: 23951
Joined: September 20th, 2002, 8:30 pm

.

October 4th, 2002, 5:35 pm

QuoteOriginally posted by: AaronTo paraphrase Herodotus quoting Solon: "call no man lucky until he's broke." We do not know what the future has in store, and recent years have seen hundred-billion-dollar piles of wealth evaporate overnight. While this scale of disaster is recent, history is full of smart and lucky people who make gigantic fortunes and then go bankrupt, often many times. In fact, it's a job requirement if you want to be a real estate developer.LOL! To respond to Aaron's paraphrasing of Herodotus quoting Solon: "I sure hope I'm not lucky!!" I keep a 20 BILLION Deutsche Mark stamp by my computer to remind me of fortunes gained and lost. It would be interesting to prove that one must choose between either mediocrity or extreme cycles of boom and bust. I suppose I seek the third way -- consistent market-beating returns at modest risk. Admittedly, it is what I seek, not what I have achieved.
 
User avatar
apine
Posts: 410
Joined: July 14th, 2002, 3:00 am

.

October 10th, 2002, 11:38 pm

i read an article by one of the guys that were working on chaos theory and markets (there was a book the predictors in which they were profiled). the article was not on chaos theory, but created a model for markets based on participant populations. the background hypothesis was similar to what traden4alpha described, ie, successful traders accumulate capital (individually or as a group) causing a self-destructive feedback loop. they took it a step further and actually modeled it. i recall them modeling "value buyers" and "momentum traders." as the author pointed out, figuring out what the market's current proportion of value, momentum and any other strategy is basically impossible. if anyone wants the citation, i can try to dig it up.