If the markets have a PREDICTABLE PATTERN then wouldn't someone be able to CORNER THE MARKET?
If the markets were REALLY PREDICTABLE then they would cease to exits.
If the markets were REALLY PREDICTABLE then they would cease to exits.
Random Walk Theory Proved, once and for all.
- Thread starter mu200411
- Start date
Quote from science_trader:
I do. Although here it's the whole concept of 'random' and 'deterministic' that is mixed. Of course most of market movements are just plain random. However, well hidden inside these movements there is an average, and if you are able to predict this average consistently well, you make money. Looking at the average backward doesn't help you. The whole question is all about what will be the average return in the future (for whatever time horizon you may choose).
How to predict it? That's your edge. If there was plain temporal autocorrelation of returns, that would be very easy. If positive, follow the trend. If negative, just be contrarian. But plain linear autocorrelation measure cannot be used as its value is well inside the statistical confidence interval. You have to find something else. And I do not even mention the problem of stationarity...
Enjoy your edge, and try to find tens of them if you want to survive
By the way, I forgot to mention that the science (or the art, call it whatever you want) of managing the "random" part of your trading is called risk management.
Edge for your deterministic returns, and risk management for the oscillations. That's all you need.
I was merely stating that random walk results do not say much about the market. The logic flaw that I noted was simple. The original poster stated that if A is random, and A can look similar to or exactly like B, then B must be random also. This is not logical, ergo the Jesus honeybun example to illustrate an equally absurd conclusion.
However, I do not believe that markets behave in a completely random manner, rather that there are so many variables of input that it may seem random when one or a few of the inputs is not overwhelming enough to dominate all others.
That's not a very good analogy. Gravity is the dominant force and is extremely predictable. You drop the leaf and you know exactly where it's going...to the ground. All the "chaos" in between is irrelevant.
I challenge you to predict the long-term trends of the market. That's where the "randomness" comes into play. The market as a whole can turn on a dime tomorrow based on macroeconomic factors changing very quickly.
The same applies to microeconomics. A company that is in "bad" financial shape can literally have a cash infusion tomorrow. Nobody without insider information can predict it. The news will be released and the price of the stock will correct almost instaneously without any forewarning. A company with problems in earnings can tomorrow gain that one contract that "fixes" their earnings problems.
It's all random. Nobody can predict it. All you can do is to reduce your risk by diversifying. If "on average" companies with good earnings outperform the market, then buy a basket of those companies. The probabilities will then be in your favor. But to say that you can predict what any one of those companies stock price will do, is pure hogwash. You can perhaps give a range of where a portfolio of those stocks will do, but that's it...it's a range.
I pretty much agree with you...
But I would use different terminology.
The key concept is "market efficiency"...
Because the degree of "randomness"...
Is more or less a FUNCTION of "market efficiency".
Highly liquid securities are very efficiently traded...
And are therefore "effectively random"...
NOT WORTH trying to exploit their small level of "non-randomness".
Many less liquid or exotic or complex securities...
Surprisingly to many... are NOT traded very efficiently...
And their higher level of "non-randomness" can be readily exploited by Quants.
Quote from mu200411:
In this thread Random Walk Theory will be proved. There will be no discussion about TA.
If I can draw charts using Random number series that look like stock charts, will it be a proof that Random Walk Theory is true or a proof that Random number series also follow the Nature law of Elliott?
Anyway let's produce some charts first.
N.B. Please move this thread to an appropriate forum.
_________________________________
Your argument that markets are random falls flat on it's face, it's easily disproved, just look at the 2 charts, one has the support & resistance zones noted after market close on 9-29-06. The 2nd chart is after market close on 10-2-06. How can this possibly be understood as random market activity? The educated trader understands why market are moving, while the uneducated trader out of frustration often times lashes out & claims that markets are random in an attempt to appease his own incompetence.
On this particular day, the sellers were exerting great influence below 11,800, while the buyers were exerting great influence above 11,720. Uneducated traders are like a rubber duck out at sea who take a pounding from the market(waves) & all they feel is pain.
http://www.elitetrader.com/vb/attachment.php?s=&postid=1219330
http://www.elitetrader.com/vb/attachment.php?s=&postid=1219331
In this thread Random Walk Theory will be proved. There will be no discussion about TA.
If I can draw charts using Random number series that look like stock charts, will it be a proof that Random Walk Theory is true or a proof that Random number series also follow the Nature law of Elliott?
Anyway let's produce some charts first.
N.B. Please move this thread to an appropriate forum.
_________________________________
Your argument that markets are random falls flat on it's face, it's easily disproved, just look at the 2 charts, one has the support & resistance zones noted after market close on 9-29-06. The 2nd chart is after market close on 10-2-06. How can this possibly be understood as random market activity? The educated trader understands why market are moving, while the uneducated trader out of frustration often times lashes out & claims that markets are random in an attempt to appease his own incompetence.
On this particular day, the sellers were exerting great influence below 11,800, while the buyers were exerting great influence above 11,720. Uneducated traders are like a rubber duck out at sea who take a pounding from the market(waves) & all they feel is pain.
http://www.elitetrader.com/vb/attachment.php?s=&postid=1219330
http://www.elitetrader.com/vb/attachment.php?s=&postid=1219331
The problem with this argument is that you are cherry picking one chart window from a universe of millions and constructing your version of what happened behind the scenes. It is a simple matter for me to cherry pick a few charts of my own, with clearly defined support and resistance levels in hindsight, then request that you tell me what the future inflection points will be (and we can say that the inflection points must have a minimum +/-% deviation from both sides of the trough or peak to be considered an inflection point). I doubt your results would be much better than determining the output of a random coin toss (notice I don't include volume and price velocity/momentum since I think these variable add significant weight to intraday guesstimates).
At best you are using support/resistance levels as a "guess"
based upon your prior experience to determine where the future turning points will lie. What's important is to determine when to get out when you are wrong in your guess. Because I can assure you from experience, that you will be wrong plenty of times.
----------------------------------------------
I've said it a dozen times, and I wish the OP would change his thesis, since it deviates from the useful argument on the trap of subjective visual interpretation of real chart patterns vs. randomly generated patterns. Markets DO NOT follow a random walk in the Gaussian sense, this has been shot down by lo, Mandelbrot and others ad nauseam.
In some ways they are better, in others worse. Better because the mean of the distribution is greater than zero (upward drift/ edge bias), worse because fat tails and large sigma variations can occur and are occasionally worse than gaussian random models predict.
Anyone who looks over my prior threads will notice I use plenty of TA in my arguments and beliefs, but I will be the first to concede that every one of those boils down to a guess whereby I believe the overall argument (TA/fund/history/probability/experience) is weighted in my favor. I believe that's the best any of us (outside of market makers with client order flow/ that's a whole nother world/debate) can do, aside from managing how we will risk our bets, and determining how we will respond to the outcome in advance.