Can linear regression analysis really predict the future?
- Thread starter tradrejoe
- Start date
there is very little to fight for I am afraid.
Each day a small number of traders place a variable number of trades totaling a large proportion of the daily volume.
If it makes you feel good to call this reality "random" then go ahead and knock yourself out.
If you prefer to call it "non random" then by all means attach yourself to this underwhelming thought.
Nothing you say or think will alter the reality of the current state of the markets.
There's research that seems to confirm what MAESTRO suggests.. I also think it's one of the coolest, as well as most famous, papers in behavioral finance, so I thought I'd attach it here.
Of course not. The same way a geologist can not change the rock formations, a biologist cannot change the way species interact etc. All we can do is to learn! The learning process however enables us to create different things that help us to adapt to the ever changing reality. It is not the reason of this discussion to "name" things (which, of course, is the subject of meronomy) but rather is to classify and group things with similar behavioral patterns (taxonomy) and learn how to utilize those patterns.
What has this laboratory outcome got to do with the reality that only a very small number of flippers control and manipulate 85%+ of the market.
I can easily grasp the isolated math of your probability example, but I think you are in need of a greater level of thought and direction when applying it to an asymmetrical object such as the market.
Anyway, as always, I will follow this thread with great interest.
Does not your statement show a lack of objectivity.
Why not give us a framework to determine whey consistent profits would over rule your theory of randomness?
By the way...
We are not questioning the existence of the rock formations - what we are question is how, when and why there might be oil stuck within those rock formations.
We engage in this silliness in order to harvest more oil.
One side says the oil just came about from a random event which caused a whole bunch of dinosaurs and trees to die over a short period of time.
The other side said after drilling thousands of holes and monitoring their oil production says - I don't know. It seems to me based on my experience there is too much oil down there to just be the result of dinosaurs and trees dying. It seems to fill up these holes in a more organized manner.
As for me my conviction may be summarized this way:
the Market is not predictable, except when it is.
Might sound like a play on words, but in fact it reflects the way I see it.
Now if only "something", be it T.A., statistical data, linear regression, splines, crystal ball ... could help me determine when the markets are in a "deterministic mood" ^^ and how to take immediate advantage of it, I'd be more than happy to agree with any theory you'd want me to.
the Market is not predictable, except when it is.
Might sound like a play on words, but in fact it reflects the way I see it.
Now if only "something", be it T.A., statistical data, linear regression, splines, crystal ball ... could help me determine when the markets are in a "deterministic mood" ^^ and how to take immediate advantage of it, I'd be more than happy to agree with any theory you'd want me to.
Rather interesting you believe that is the "only" reliable way of making money in the markets. This is an increasingly common "quantway or the highway" attitude that is both arrogant and misplaced. Tell that to the 1000's of successful traders before and after the dawn of ARCH mingling with GARCH @ the CoIntegrated Surface Distribution party.
Professionals who tend to be highly invested in their intelligence, particularly self-esteem wise, tend to be the most rebellious. Many come to realize that basic time-proven trend following behavior completely annuls their need and ability to understand complicated econometrics, including the emotional need to develop esoteric models that attempt to show why equities/commodities go up,down or revert to a mean. That typically doesn't feel good as there is an inherently strong desire for appreciation.
Basic time-tested trend following refers to massively plain templates such as a SP500 leveraged long-short 50SMA-daily/200SMA-daily interaction @ 1.0 Kelly. Overwhelming probability is that ~80% of quantitative-based HF/FoF's will be unable to outperform this "5th grade math" position trade approach over the next 10yrs, just as that same % was unable to do so over the past 10yrs. Now should that overshadow the highly effective/superior performance of that remainder ~20%? Absolutely not, only a sobering perspective that Q's IMHO need to appreciate and show humility towards. Your method is one way "you" have uncovered to remain consistently profitable up to this point... a method in a sea of other successful approaches. To state it is the only approach is patently absurd.
....2c from a fellow enginerd who uses the likes of QH QuantDeveloper&DB/MatLab FT/SPSS/R etc on a near daily basis.
Interesting thread. One of the more stimulating threads on ET in some time. Approaching NP/Wilmott grade ...recognize a few NP members posting in here
i think what Maestro is trying to say here is that a strategy has a real edge only if a few 1,000s of people apply the same strategy and most gets a +ve performance, while some other strategies used by some successful traders if applied by a few 1,000's of people and most may not get a +ve return and so to conclude those other strategies actually do not have an edge and that those traders are successful so far is a result of randomness as what he mentioned sort of like taking millions of investors and one will turn out to be a Buffet, am i right? so the only way to profit is to narrow down into high freq trading so as to isolate as many external factors as possible as the more time is considered in case of a few days to weeks or months, the more unpredictable factors are involved, sort of like a closed box experiment in labs thing