Intuition Amplifiers 2

All it says that if you flip a fair coin and create a random walk out of the multiple flips you will have a chart that is based on a normally (Gaussian) distributed data. It has been proven beyond any doubts that no matter what tactic you would use to trade this chart you guarantee not to win (and not to lose either with the exception of commissions and Bid/Ask spreads). The only way to make money on the markets (aside being a broker or a bank) is by relaying on the market inefficiencies. Those inefficiencies exhibit themselves as stable differences between the normally distributed data and the data that has “non-randomness” built in to it. Once those inefficiencies are recognized and extracted they have a tendency to be “arbed out” of the market. It is incredibly hard to find any “stable” inefficiency. So, majority of people do not bother to check whether their TA tools really can find those market inefficiencies (differences between the normally distributed data and the distribution skews) or not. That is why when I was dealing with Bank of America back in 2007 the first test they asked me to perform is to show what would happen to my methods if they are exposed to totally artificially created, random data. Any TA indicator, any chart tool anything at all must be tested against those conditions. If they work the same way on normally distributed, random data as they do on real charts that simply mean that those tools are no good at all. Only when it has been thoroughly proven that there is noticeable difference between their reactions to different data origins you can accept those tools as being genuinely efficient. Unfortunately 99.9% of all known to me TA methods and tools fail this test.
Cheers,

MAESTRO

 

You are too smart for me But isn't that true that without your excessive knowledge and your hands-on experience you wouldn’t be able to tell? This is exactly what I am trying to say. This is one of the most powerful demonstrations of INTUITION. Thank you BTW, you are as always bang on!

Cheers,
MAESTRO

 

we remain positive for the year due to the 600 point, 10 contract killer trade at the start--yeah, sucked lately, but March isnt over yet...

surf

 

I think your ps post is confusing. There are certainly winners and losers after a distribution has happened random or otherwise. Look at annual (past) returns of hedge funds, the problem is what is the future return and its variability (vol). I know what you are saying though, I would say that there are streaks in coin flips that can be ridden/exploited.

http://www.tvmcalcs.com/blog/comments/coin_tosses_and_stock_price_charts/

 

If the returns were drawn from a normal distribution, I wish I could tell in advance the order in which the returns would be drawn.

For the above -- so this is why my backtesting (when I used to do it) sucked so much. I followed the usual approach of develop on some recent past data, then apply to other sets of past data for comparison, then trade forward on new data as it came out. The performance in all 3 types of data sets had no relationship, and forward testing always performed like a random system (system, not data).

But if I understand above, a better process would be as: have 10 sets of real data; have 10 sets of randomly generated data, presumably normally distributed; test hypothesis: strategy can determine which of the 20 data sets are real vs. random; statistics to determine if it chose with results significantly greater than 50%. (I'll have to look up a good statistical test for this, or check the significance levels.) Or make it tougher, with 5 sets of real data vs 95 sets of artificial random data.

 

here is a chart created randomly, it is a possible distribution, is it tradeable?. I bet most TA folks would say yes. Is it THE future distribution, no of course not.

 

I betcha nodoji, cornix, hershey and the other hindsite biased deluded souls would say they would make money on that chart. surf

 

Well, let's kindly ask the Xspurt. I think you would be amaized by his answer.

 

Would you ba able to put up the independent variable of the market?

All you show is the dependent variable.

Most science and math based systems use the independent variable to take the full offer of the dependent variable.

TIA.

 

I went to the science museum and there was a mock up of a model river and the challenge was to build a bridge over it with the wooden blocks provided. They had lines drawn to say building to this span is ok, then further it was good, more was excellent and the final line was impossible.

I started and completed the impossible, not that it was hard. I wondered if they were filming to see how many people would accept the mental confines of the lines.

The market is like that. I never accepted the impossible was a correct verdict so I invented my own TA. But isn't that exactly what you are doing - proving what others laughed at a few years back? If I didn't have the tools then I still wouldn't accept the premise that the market is random because I am bloody minded when it comes to my intuition and it starts with doubting failure. Someone elses failure is not mine until I make it so.

To answer your question, no without the tools I wouldn't be able to see the answer, even when it is staring me in the face. But somehow, sometimes, you just know that there is something there asking you to look deeper. So it starts with intuition, moves to understanding and rules of logic and then back into intuition when everything speeds up.