Why do I see "Trends" in Randomly Generated Data?
- Thread starter Rahula
- Start date
I gather that you subscribe to the principle of reversion to the mean. (Please correct me if I'm wrong.) If so, then how is the expected reversion not an exploitable trend, as another poster mentioned earlier and as I had alluded to in other threads?
Behavioral Finance:
A field of finance that proposes psychology-based theories to explain stock market anomalies. Within behavioral finance it is assumed that the information structure and the characteristics of market participants systematically influence individuals' investment decisions as well as market outcomes.
A field of finance that proposes psychology-based theories to explain stock market anomalies. Within behavioral finance it is assumed that the information structure and the characteristics of market participants systematically influence individuals' investment decisions as well as market outcomes.
You are one complex carbohydrate, aren't you?Quote from MAESTRO:
No, I'm sorry; it is not what I subscribe to. Reversion to the mean as any other technique related to price patterns is not in my tool box. I thought that the article I posted would give you a clue...
Sorry, I have not yet had a chance to read your article.MAESTRO,
What can I say. Medalion successfully trades around $4 billion by utilizing intra day models. Some of which were revealed during the lawsuit when 2Phds left the fund. One was to deal with gaps and another with analyzing level 2 data. Their models utilize repeating patterns (pattern has many definitions).
Regards,
redduke
I'm not into any random testing unless all data feed variables are involved and many fractals are deployed.
But when I do read and explore efforts, I do see that a lot of people approach randomness from the viewpoint that anything is possible. I know there must be some work done on the 101 level that convinces some people to not do any pattern work. I am also sure that this work is rejected by others who then look at patterns in randomness.
French, et al appealed to me (not the momentum studies but the results in terms of locales) because I have been using Universes forever to maximize the trading gains by choosing universes that make the most money. They happen to be momentum universes and not those chosen by big money which can't make money (as shown by their results so far).
I was glad to find out the thrust of M's work efforts; it certainly clarifies the the why and the way he is working at getting to any conclusions.
Interesting theory. It's funny how I've often thought about the similarities to market modeling and philosophy. They are very similar. You have the die hard platonists that believe there is an underlying objective reality behind all, then there are the existentialists who are horribly depressed by realization that it is all rather chaotic and there really is no underlying order, they've come to such realizations as religion is simply a fabrication (ex TA cults?).
Nietzsche once said something to the effect that no matter how deep science reaches and how minuscule their microscopes, they will never uncover an absolute truth. Rather like icarus, who flew to the sun, at best, they might be blinded. About the only thing I can say about the holy grail, is that once the masses find it, the rules will change and the grail will disappear (SOES anyone?). Sorry for the digression, rambling out loud.
BTW. Although I didn't read it in detail, your paper seems to be saying very similar arguments to what I mentioned about a positive offset in the mean of the markets. Great work. You seem to use a lot of engineering signal terms in your work (omega, non-zero crossing, bias)... just an observation.
There's a great layman's book that has been out for awhile. Reading your background, I think you would get a kick out of it. It's called "origin of wealth," by beinhocker.
http://www.amazon.com/Origin-Wealth...bs_sr_1?ie=UTF8&s=books&qid=1204064679&sr=8-1
The book attempts to explain how modern physicists are trying to rewrite the classical rules of economics using complexity, and non-linear dynamics, as well as game theory, and behavioral psychology as tools in their arsenal.
I particularly like the studies they model whereby once a set of agents stumbles upon a successful trading rule, it quickly takes off then dies as more and more agents jump on. While that seems obvious in hindsight, the objective simulations they ran on it were pretty interesting.