Why do I see "Trends" in Randomly Generated Data?

Quote from Jerry030:

I would agree, adding that any technology sufficiently advanced appears as magic or as impossibility to those who lack an understanding of it. Einstein made this point around the time traditional physicists were scoffing at his new and unproven theories.

Perhaps this debate on are market randomness has a similar division. Some take a definition of random and attempt to find ways that the market matches that definition and assume this proves something about the market. It only proves that given enough motivation one can find arguments to support any proposition in most domains.

Take the argument over say the surge in Iraq. Are large bombings down? - Yes. So is it the solution for that mess? If you just count bombings you have to say yes. Are the forces that control the Iraq government using it to build a stable government that addresses the religious divisions that are hundreds of years old?...No. So short of keeping US forces there for 100 or 500 years to suppress the civil war, will the surge bring peace and democracy? Unknown, but I wouldn't invest in it.

A suggestion: for a moment let’s abandon theoretical arguments on what the market is.

Instead will those who advocate a random nature state a practical measure which if exceeded would cause them to question their position? Those with a non-random perspective can do the same.

Random guys might say:

A Profit Factor greater than X for at least Y trades over Z bars would call into question the random theory.

Non-Random guys can counter:

Fine, here is an approach that exceeds those numbers.

Thus we go from theoretical debate to a "boots on the ground" actuality.

Jerry030
More...



this is sort of twisting reality.

the definition of market order is ALWAYS based on the individuals viewpoint. " i can see the order" " i can profit from seeing the order" and obviously " the order, that i and only i see, allows me to profit"

the randomness crowd looks at the market from a detached viewpoint generally-- the market as a whole, not individually based.

it's the SUBJECTIVE v. OBJECTIVE dilemma----- the subjective seeing of order is meaningless to the objective observer.

and NO your profit from thinking you see order thereby predicting the market--- has no bearing on the subject. you're the outlier to the whole.

surf
 
Quote from Jerry030:

Instead will those who advocate a random nature state a practical measure which if exceeded would cause them to question their position? Those with a non-random perspective can do the same.

Random guys might say:

A Profit Factor greater than X for at least Y trades over Z bars would call into question the random theory.

Non-Random guys can counter:

Fine, here is an approach that exceeds those numbers.

Thus we go from theoretical debate to a "boots on the ground" actuality.

Jerry030
More...

Nice idea but that would require original thought from both sides and that is something impossible to accomplish thus far in these hollowed pages.
 
Humans typically explore the universe from their ego stand point of view. It is virtually impossible for an average person to imagine that the collective behavior of many target oriented, highly knowledgeable and intelligent people is as random as the Brownian motion of gas molecules! If a 1000 people would be asked to trough darts into a dart board the overall density pattern of darts around the central circle would be very close to the Gaussian distribution curve. Despite the fact that each and every person would do his/her best to aim the darts they will still lay in this magnificent pattern! And nothing could be done about it! The link below is demonstrating this process using a very interesting applet.

http://demonstrations.wolfram.com:80/StatisticalMechanicsOfMoney/
 
Quote from MAESTRO:

Humans typically explore the universe from their ego stand point of view. It is virtually impossible for an average person to imagine that the collective behavior of many target oriented, highly knowledgeable and intelligent people is as random as the Brownian motion of gas molecules! If a 1000 people would be asked to trough darts into a dart board the overall density pattern of darts around the central circle would be very close to the Gaussian distribution curve. Despite the fact that each and every person would do his/her best to aim the darts they will still lay in this magnificent pattern! And nothing could be done about it! The link below is demonstrating this process using a very interesting applet.

http://demonstrations.wolfram.com:80/StatisticalMechanicsOfMoney/
More...

Yes, true but the point of trading the markets is to make money at a rate above the average common available return (or random if one wants to call it that.)

That goal has nothing to do with an appearance that a person calls by any particular name or what they say their intention was. If a given person can hit the smallest circle in the bulls eye 9 times out of 10 over 1000 attempts in contrast to the average of all other participants who do so .01 times out of 10 over 1000 tries, we can say their skill in hitting the bulls eye is very exceptional.

You can call it luck, they can say it is skill, another person can say its divine intervention or that they can do so because they are actually a visitor from another planet disgusted as a human.....many theories can be offered...BUT the fact remains that their accuracy is exceptional.

A debate on semantics may have some academic or entertainment value but from my perspective the debate isn't likely to produce much profit. I trade for profit. The questions of importance from that perspective are:

1) Can exceptional profit be extracted from the markets?

2) How can that be done?

I'll take a stab and defining exceptional profit:

Profit Factor: > 3.5
Number of Trades: > 200
Ratio of Bars in Trades to Total Bars in Time period: < 10.0

What do you think?

Jerry030


FOr example the retrun on my IRA is better than the prime rate but nothing that one could call an exceptional rate of return on my money.
 
Quote from Rahula:

I've created an excel file I call "random chart generator". The weird thing is that most of these charts look exactly like market charts and you can apply all kinds of TA tools like moving averages, and trendlines to these charts.

If you don't believe me, see for yourself. Type in =RAND()*(-1-1)+1 into excel in column A. Then create a rolling sum of column A in column B. Then create a chart based on column B values and add a moving average.

Now what to make of this? I suppose when most people see a 7,7,7,7,7,7 result in the lottery they see a 'trend' even though it is just as random as 23,42,11,14,33,8 lottery result - both events have an equal probability of happening.

Now if the markets are random and the market closed up,up,up,up,up,up,up,up,up,up,up,up 12 days in a row - most people would likely be talking about it as a n-sigma event or a "black swan" even though it is just as unique and random an outcome as a up, up, down, up, down, down, down, up, down, up, down, down 12 day series of closes.

Now doesn't a random market assumption mean that both mean reversion and trend trading strategies don't have any science to back it up and at the end of the day it all comes down to luck?

I know, scary thread.
More...


what you are describing is the same paradox in mathematics that occurs when one attempts to computerize and automate what used to be paper calculations and HP Scientific Calculator equations done in long form, or simply written out.

what you are also describing is the same thing that programmers tasked with in generating testing data to apply towards their programs, and using random generators have observed,,

in both cases, the conclusion is very similar to what you stated in your premise, that these random number generators and random data generators are not that random in their originality or creation of said data sets...

now, what lacks credulity, it the notion that one can then apply this phenominon to real life trading and real life data collction...
 
Quote from marketsurfer:

commonly available return = random ??!
:eek:



no offense dude, but this book should offer some guidance:
http://www.amazon.com/Innumeracy-Mathematical-Illiteracy-Its-Consequences/dp/0809058405
More...


In terms of your question "commonly available return = random "

I was trying to get out of the academic and theoretical into the practical. Labels and what they mean to each person is often a hindrance to practical communication and real world results.

So for purposes of my question lets assume that there is a common rate of return on investment activity. If it is impossible to trade the market with any predictive ability them by definition the results are random if the market is random. Regardless of the random or non-random nature of the markets, if we take several million individual investors, a few thousand mutual and hedge funds and average their returns we have the average or commonly available return. I suspect this is somewhere between 0 and 10%, but that is just a guess.

Would you agree?

Would you agree that a group or method that can get a 50% or 100% return is not common or average?

This is independent of the definition of the market as random or non-random...just what is the average and non-average return on account.

Thanks for the reference to the book but the lack of mathematical sophistication of the American populace has little to do with making money in the market. It all just labels and terms being tossed about with no intrinsic meaning, at least in terms of profitable trading.

In the Dark Ages Catholic theologians used to debate how many angels could dance on the head of a pin. Any angels observing the discussion probably laughed themselves silly since there probably was more value in seeing an angel or talking to one than debating their size. IMO, the value is making money not debating the best label to attach to the market from theoretical terms.

So the question is better framed through the following logic:

a) What level of profitability would indicate non-random investment performance?

b) Does that level exist?

c) If it does then the market is non-random.

d) if is does not then until it exists the question can’t be answered with certainty

200 years ago I'm sure physicists debated if it was theoretically possible to fly like a bird. They likely gave up the debate when some bicycle makers form Ohio invented the airplane in 1903.

Jerry030
 
Quote from marketsurfer:

this is sort of twisting reality.

the definition of market order is ALWAYS based on the individuals viewpoint. " i can see the order" " i can profit from seeing the order" and obviously " the order, that i and only i see, allows me to profit"

the randomness crowd looks at the market from a detached viewpoint generally-- the market as a whole, not individually based.

it's the SUBJECTIVE v. OBJECTIVE dilemma----- the subjective seeing of order is meaningless to the objective observer.

and NO your profit from thinking you see order thereby predicting the market--- has no bearing on the subject. you're the outlier to the whole.

surf
More...

Well yes, if it's human trading or looking at the market. If you have a non-human, say a neural network doing the looking the question is can it learn something that it can use to predict future market activity?

The nice thing about the more sophisticated of these is that they turn what they learn into predictions which can be tested against the future and a conclusion made based on their ability to generate profit. This eliminates the subjective entirely since the NN has no emotions about its performance or what is sees or learns at all.

Jerry030
 
i see this argument over and over thru all posts in several differ,ent journals,there never is an answer, its like philosophy, if we base our argument on an if rather than an is, we never are sure of our results,i must complemrnt this thread as even tho no ones sure, no one has pre tended to be sure with an insult to make thier unproven point, carry on in your quest to quantitively figure out the mrkt,if you get an answer,we'll move on to more complicated subjects, women,logic,god ,longevity,health, all the answers are here, it's just a question if they are right or just raise another question, if its gonna rain and you are working outside , do you get a half hour notice? usually u do....the mrkt is the same way ..if you just stand back and look you usually know something is not rosy, trust your feeelings here.. go with it..there is some proof always but if you go with what you feel ,not think, its always there to warn you

/
 
Quote from ammo:

i see this argument over and over thru all posts in several differ,ent journals,there never is an answer, its like philosophy, if we base our argument on an if rather than an is, we never are sure of our results,i must complemrnt this thread as even tho no ones sure, no one has pre tended to be sure with an insult to make thier unproven point, carry on in your quest to quantitively figure out the mrkt,if you get an answer,we'll move on to more complicated subjects, women,logic,god ,longevity,health, all the answers are here, it's just a question if they are right or just raise another question, if its gonna rain and you are working outside , do you get a half hour notice? usually u do....the mrkt is the same way ..if you just stand back and look you usually know something is not rosy, trust your feeelings here.. go with it..there is some proof always but if you go with what you feel ,not think, its always there to warn you

/
More...

I I assume you are a discretionary trader?

If you trade a mechanical system there are is no looking or feeling involved, only outcomes. My system trades without any human intervention at all. I could spend a month camping in the wilderness without any communication and as long as my PCs had electric power and an Internet connection it would continue to day trade the markets.
 
You must log in or register to reply here.
Back
Top