Determining randomness in a trading system | Page 2 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Successful Trading & Determining the Probability of a Trading System 10 replies

Fooled by randomness 15 replies

Luck, Randomness, and Chance in Trading 0 replies

Thoughts on randomness and systems 40 replies

Poll: Are trends rooted in randomness? 16 replies

Attachments: Determining randomness in a trading system
Exit Attachments

Determining randomness in a trading system

  • #21
  • May 30, 2012 1:08pm May 30, 2012 1:08pm
Quoting Skynet
Fascinating.
From reading the market wizards and market histories...i am of the opinion that : A technical/fundamental approach CAN ONLY BE solidly functional for a specific kind of market within a certain period of time(+/- 10 years...if you are lucky...15yrs). After the 10 year mark, the trading approach will start to fail unless it has undergone updating to reflect the new market reality. The trader that hasnt done this will start to see his method fail.

[b]Logically, i dont think you can find a method that works in all market condition....
a little unrelated, but still important in my opinion:

what I also like to do with my testing is to find out about the predictive value of my systems. Say I backtest and optimize my systems from beginning till 2009. Now I look at 2010. If the profits of 2010 are lower than in the year 2009, then my system has a low predictive value, if it is higher than 2009 then it has a high predictive value.

That was just a simplification. I do that for all years from the beginning till 2012. Make a ratio for each year (predicted year (i.e. 2010)/last year of the optimization period (i.e. 2009)) and look if the average of all these ratios is equal or higher than 1. If it is not, then back to the drawing board. That weeds out a lot of "promising" systems, that actually are either overoptimized or just had some good streaks but go back to random.
  • #22
  • May 30, 2012 1:51pm May 30, 2012 1:51pm
Quoting Skynet
Why is it that GANN'S method worked so well when it did? Why is that elliot wave method worked so well, when it did? why did breakout used to work like a charm? same with daily pivot points? what was going on in terms of fundamentals/sentiments/types of market players when those methods were very brutally effective? What kind of market was it?
Easy, as long as a lot of market participants aren't aware of a particular strategy the same one will work. Why? Because alpha returns are always zero sum by definition and you can't have majority of participants profitable. Fx volume is 99% speculation and not held for delivery so it's a zero sum game; taking money out of someones else account - mostly banks since they'll take the other side of your trade....

People that make winning systems don't post them on public forums like ff or elitetrader,
or if they do they found that in a year span the game is a lot tougher for that particular strategy,
this is why only 2% people make money trading FX in the long run,
because it's only 2% that make money and shut up
  • #23
  • May 30, 2012 2:13pm May 30, 2012 2:13pm
Quoting Rikers
Easy, as long as a lot of market participants aren't aware of a particular strategy the same one will work. Why? Because alpha returns are always zero sum by definition and you can't have majority of participants profitable. Fx volume is 99% speculation and not held for delivery so it's a zero sum game; taking money out of someones else account - mostly banks since they'll take the other side of your trade....

People that make winning systems don't post them on public forums like ff or elitetrader,
or if they do they found that in a year span the...
1. Trading is often as much about common sense as it is a rigidly defined system (manual trading)?
2. Trading forex is not a zero sum game.
3. Trading isn't about the majority of participants, it's about where the majority of the money is.
4. To say that something will become unprofitable once it becomes known is a bit like saying McDonald's Hamburgers is now defunct because everyone knows their model.
When the Joker is in the deck.. fear not and play it well
  • #24
  • May 30, 2012 2:30pm May 30, 2012 2:30pm
Quoting Rikers
Easy, as long as a lot of market participants aren't aware of a particular strategy the same one will work. Why? Because alpha returns are always zero sum by definition and you can't have majority of participants profitable. Fx volume is 99% speculation and not held for delivery so it's a zero sum game; taking money out of someones else account - mostly banks since they'll take the other side of your trade....

People that make winning systems don't post them on public forums like ff or elitetrader,
or if they do they found that in a year span...
Pay attention, traders. This post is one of the most sensible and true posts on this entire forum.
  • #25
  • Edited 4:34pm May 30, 2012 3:48pm | Edited 4:34pm
Quoting Rikers
Easy, as long as a lot of market participants aren't aware of a particular strategy the same one will work.
"avalanche" phenomenon is not the only reason why a method fails to work or not work. Fundamentals play a role. Market sentiments play a role.

Take gold or tech bubble or tulip bubble or housing bubble. Everybody and their grandmother were in it and yet they kept going higher and higher and higher. If you were using breakouts or SMAs during that time...or classic turtle approach, it would have worked, it doesnt matter how many people knew about breakouts or how many people bought simply because price dipped(into the 100 or 200 smas) or macd bullish divergence. All those methods would have worked. A detail statistical study of various technical patterns under different market conditions was done by thomas bulkowski in "encyclopedia of technical patterns.".

During the 2008 economic meltdown with massive deleveraging and unwinding of carry trades. Did breakouts not work? Of course, it did. It didnt matter how many people were aware of breakouts methodologies.

TO mr slim buffet and rikers,

On whether majority participation affects price behaviour or not. I can only point in the direction of these academic studies:
(perhaps, redlion is familiar with the GARCH model?)

#1. Price Formation in Financialized Commodity Markets:the role of information.

#2. Avalanche dynamics and Trading Friction effects on the Stock Market.

#3. Empirical relationship between trading volume, returns, and volatility.

On top of that, there has been studies done on the effect of algorithmic trading on the market.

Of course, some will argue stock market vs commodity market vs forex market.
" It may mean something. It may mean nothing at all "
  • #26
  • May 30, 2012 4:00pm May 30, 2012 4:00pm
Hi tnx for the post,

I didn't argue against majority participation affects but you can in essence look at every possible trading logic as a separate strategy - even the fundamental players would "arb" up the profitable moves if there were enough volume from them.

Bubble is not a phenomena once in a while but a usual market state if you trade on M5 period. There is no special market state that is exhibited in a year before that isn't on a month of trading on M5 chart.

If you have a link for studies done on the effect of algorithmic trading on the market please pm if you would. Sounds interesting.
  • #27
  • May 30, 2012 4:02pm May 30, 2012 4:02pm
Quoting Skynet

TO mr slim buffet and rikers,

On whether majority participation affects price behaviour or not. I can only point in the direction of these academic studies:
(perhaps, redlion is familiar with the GARCH model?)
I'm sorry but I really don't have time to read through all that. I can only say that I prefer to define a "method" and a "system" as two different things.

As with a successful franchise, I try to take something good, improve upon it and change with the times as need be.

To your success.
When the Joker is in the deck.. fear not and play it well
  • #28
  • May 30, 2012 4:05pm May 30, 2012 4:05pm
Quoting Skynet

Logically, i dont think you can find a method that works in all market condition.

my two cents.

You don't find a SYSTEM that works in all market conditions.

proven
When the Joker is in the deck.. fear not and play it well
  • #29
  • May 30, 2012 4:31pm May 30, 2012 4:31pm
Quoting Roofx
If you believe that the most successful traders go around posting their methods on forums so all can be winners then your chances of succeeding are true to your user name. No offence.
If you believe that successful methods remain unknown then you are
UNKNOWING

fairly simple
When the Joker is in the deck.. fear not and play it well
  • #30
  • May 30, 2012 4:34pm May 30, 2012 4:34pm
Quoting Slim Buffett
4. To say that something will become unprofitable once it becomes known is a bit like saying McDonald's Hamburgers is now defunct because everyone knows their model.
Ridiculous comparison. You have taken the words out of context and applied them to a completely different subject matter. We are talking about trading foreign exchange, not selling burgers. A better comparison is that of two athletes. One competes against the other to take the prize out of the other's pocket. One day Athlete Bob asks Athlete Sam "how come you always get to the finish before me?" Athlete Sam replies "Oh, it's because I have a method that you don't know. You see, I take vitamin Z mixed with beetroot 3 times per day and it makes my muscles much more powerful and my stamina greater". Next week Athlete Bob begins the same suppliment intake and wins the race.
  • #31
  • May 30, 2012 4:37pm May 30, 2012 4:37pm
Quoting Slim Buffett
If you believe that successful methods remain unknown then you are
UNKNOWING

fairly simple
Successful methods remain unknown to the majority of traders. Once everyone is using them they no longer work. If everyone used the same winning method correctly how could everyone make money? Where would all this extra money come from? Space? The clouds? When one trader makes a profit that profit comes from somewhere, not some magical never ending money pit.

What worked 20 years ago no longer works in its unchanged form. The reason we have to keep evolving our methods is to stay ahead of the competition.
  • #32
  • May 30, 2012 4:45pm May 30, 2012 4:45pm
Quoting Roofx
Ridiculous comparison. You have taken the words out of context and applied them to a completely different subject matter. We are talking about trading foreign exchange, not selling burgers. A better comparison is that of two athletes. One competes against the other to take the prize out of the other's pocket. One day Athlete Bob asks Athlete Sam "how come you always get to the finish before me?" Athlete Sam replies "Oh, it's because I have a method that you don't know. You see, I take vitamin Z mixed with beetroot 3 times per day and it makes...
Nice little story.
In the 1950's a franchise was started. They had a particular way of doing things. They improved upon the method and changed with the time.

The "athletes" of the franchise are long since dead but the business lives on.

helluva deal !
When the Joker is in the deck.. fear not and play it well
  • #33
  • May 31, 2012 1:43am May 31, 2012 1:43am
I belive that some systems can hold up over time given that they are robust enough. Earlier in the thread an example of a MA crossover system was mentioned.
To me that is an example of that.

Sure you will have periods of drawdowns that can stretch for years but eventually it will go back to a profitable state.
Swedish steel!
  • #34
  • May 31, 2012 6:33am May 31, 2012 6:33am
I see the thread has been heavily moderated and many posts have been deleted.
  • #35
  • May 31, 2012 6:37am May 31, 2012 6:37am
Quoting grovsnus
I belive that some systems can hold up over time given that they are robust enough. Earlier in the thread an example of a MA crossover system was mentioned.
To me that is an example of that.

Sure you will have periods of drawdowns that can stretch for years but eventually it will go back to a profitable state.
In my opinion MA crossover systems are for so called teaching schools to "teach" to their students. Moving Averages cross over because price moves. A moving average crossover says "hey look, a while ago price moved". We don't need a MA cross to show us that price moved X amount of time ago, we can see it with our own eyes. Like the vast majority of indicator based systems, they can make profits in good trending conditions, but in ranges they get thrashed.
  • #36
  • Last Post: May 31, 2012 7:11am May 31, 2012 7:11am
Quoting Rikers
Hi tnx for the post,
......
Bubble is not a phenomena once in a while but a usual market state if you trade on M5 period. There is no special market state that is exhibited in a year before that isn't on a month of trading on M5 chart.

If you have a link for studies done on the effect of algorithmic trading on the market please pm if you would. Sounds interesting.

I wrote a detailed reply with examples...just before i posted it, my browser suddenly close on me....i just didnt feel like re-writing that post again.

i hate it when that happens.
" It may mean something. It may mean nothing at all "
0 traders viewing now