Menu
Sister Sites

Options

Bookmark Thread

First Page First Unread Last Page Last Post

Print Thread

Similar Threads

Is Grid Trading (combined trend following and counter trend) Profitable? 3 replies

Making a good diversified portfolio 10 replies

Diversified EA Money Management Code 10 replies

HAMA PAD - A Simple Trading Approach 7 replies

Trading a (semi) automatic approach...but want more 11 replies

Attachments: Diversified Trend Trading Approach
Exit Attachments

Diversified Trend Trading Approach

Fear not thy apparent dabbling with convergence....for ye have little faith. It still manages to retain some of it's divergent qualities and prevent it's full demise into the land of negative skew......so perhaps I am just a little pregnant with the convergence in it.

Yeah though I walk through the valley of convergence I fear no evil.

As you can see below...any illusory edge is disappearing before mine eyes with more long range data under the belt. My divergent resolve was tested in the desert....but I will return to my holy place.

5 instruments down and the weakness of the strategy is shining though...not merely as a poor divergent solution...but even as an auto-correlated steroid.

Looks like it is going to be a fizzler.

Note the strong performance over the short range testing period that got me excited.

Not that your idea is not worthy Pip....but this system is weak in every measure. Despite the looming failure with this system....I will ensure that all future tests I do also test for auto-correlation of the equity curve...just in case future testing brings out this principle again which might be harnessed.

Attached Image (click to enlarge)
Click to Enlarge Name: Capture.JPG Size: 74 KB


It's not over yet...but I can hear the distant cries of the fat lady!!!!!

All is not in vain. At least Tickstory2 is getting a good workout during it's 14 day demo period.

Please keep a divergent spare seat for me as I will be returning shortly it seems.


C
How harvesting large move looks like using EDTT and KAMA. Kama with 4h filter and local high break.

Short term EDTT is top earner.

Karma:
Attached Image (click to enlarge)
Click to Enlarge Name: biLfZ4h.png Size: 114 KB


EDTT:
Attached Image (click to enlarge)
Click to Enlarge Name: 4BiuYPg.png Size: 160 KB
Cut short your losses. Let your profits run on. David Ricardo (1772-1823)
Quoting Copernicus
Fear not thy apparent dabbling with convergence....for ye have little faith. It still manages to retain some of it's divergent qualities and prevent it's full demise into the land of negative skew......so perhaps I am just a little pregnant with the convergence in it. Yeah though I walk through the valley of convergence I fear no evil. As you can see below...any illusory edge is disappearing before mine eyes with more long range data under the belt. My divergent resolve was tested in the desert....but I will return to my holy place. 5 instruments...
Large sample size rules and save us from the damnation of illusory hell.

Guys from Strategy Quant recommend to stick to 1h time frames and higher for algo trading....they say less than this is not reliable. So let's start testing from 1h up also with autocorellation?
Cut short your losses. Let your profits run on. David Ricardo (1772-1823)
Nice explanation of 4th dimension. Tesseract is possible to imagine (barely), but how cube of 5th dimension would look like is too far. Then there is this theory of E8...

Inserted Video
Cut short your losses. Let your profits run on. David Ricardo (1772-1823)
Human nature is why we have trends and here is first experiment showing it:

Inserted Video


https://en.wikipedia.org/wiki/Asch_c...ty_experiments

75% and 1% is quite a difference.

Watch for this bias....
Cut short your losses. Let your profits run on. David Ricardo (1772-1823)
Quoting Ramadas
How harvesting large move looks like using EDTT and KAMA. Kama with 4h filter and local high break. Short term EDTT is top earner. Karma: {image} EDTT: {image}
....and here we are back to that old trade efficiency idea of EDTT with the euclidean geometry and no curves. After sleazing around with the new good looking girlfriend (KAMA) with the great curves you forget the value of your older less endowed wife (EDTT) :-)
Quoting Ramadas
{quote} Large sample size rules and save us from the damnation of illusory hell. Guys from Strategy Quant recommend to stick to 1h time frames and higher for algo trading....they say less than this is not reliable. So let's start testing from 1h up also with autocorellation?
Will do R. I must admit this short term land scares me a bit. So many critters and stuff that bite and sting.

Just gotta finish off the test on auto-correlation on M5 as a matter of principle. Will post the final results on this thread as a warning and reminder.....or even perhaps a sort of nice surprise.
Quoting Ramadas
Human nature is why we have trends and here is first experiment showing it: https://www.youtube.com/watch?v=NyDDyT1lDhA https://en.wikipedia.org/wiki/Asch_c...ty_experiments 75% and 1% is quite a difference. Watch for this bias....
Keep em' coming R. Great stuff :-)
If you have time to ponder on this concept - Auction Market Value Theory & Analytics Expert Advisor - https://www.forexfactory.com/showthread.php?t=772227
Quoting Ramadas
Nice explanation of 4th dimension. . . ..
Attached Image


It is about price boxed in a time frame.
Quoting failinforex
If you have time to ponder on this concept - Auction Market Value Theory & Analytics Expert Advisor - https://www.forexfactory.com/showthread.php?t=772227
Cheers F

I just had a quick squiz and it could be interesting. The EA might be a useful addition. What he is experiencing is the same as us. You don't need much of an edge with a system to push out some severe performance results under wide diversification.
Quoting Copernicus
{quote} Will do R. I must admit this short term land scares me a bit. So many critters and stuff that bite and sting. Just gotta finish off the test on auto-correlation on M5 as a matter of principle. Will post the final results on this thread as a warning and reminder.....or even perhaps a sort of nice surprise.
I will stop backtesting now. I am tired.

Attached Image (click to enlarge)
Click to Enlarge Name: Capture.JPG Size: 80 KB


The issues of the Auto-correlated EA (steroid) on the KAMA crossover shines through with the data.

For a start... we have 22,544 trades for the analysis which is great as it is putting the illusion to the test.

This is not a solution for all instruments. Only some instruments. That is a red flag to a bull for me....A robust edge under diversification means that we need to be agnostic regarding the liquid instruments we choose. To select AUDCAD or EURCAD in this instance would be cherry picking.

Attached Image (click to enlarge)
Click to Enlarge Name: Capture2.JPG Size: 164 KB


With 9 instruments tested and only 2 showing occasional signs of auto-correlation..this solution will go to the sin bin of lost EA's. There is not enough muscle in this system to meet the robustness test.

Attached Image (click to enlarge)
Click to Enlarge Name: Capture3.JPG Size: 103 KB


Now just imagine you were trading AUDCAD over the last couple of years. You would think that you had found the holy grail....however the result clearly demonstrates why you need to throw significant data at the solution you are testing spanning an array of different market conditions to sniff out the weakness in your creation. Market conditions can persist for years......and then suddenly change on a whim...For example refer to EURCAD for the period 2013 to 2014.

Back to the divergent drawing board. :-)

C
Plato's Allegory of the Cave
Quoting Copernicus
I will stop backtesting now. I am tired.
Attached Image (click to enlarge)
Click to Enlarge Name: The-Allegory-of-the-Cave-by-Plato.png Size: 856 KB
Quoting CrucialPoint
Plato's Allegory of the Cave {quote} {image}
Hehehe now you're just being nasty CP :-)

Let me live peacefully in my realm.
Quoting CrucialPoint
Plato's Allegory of the Cave {quote} {image}
Traders are like the prisoners in Plato's cave , they see only parts of the world that are apparent as they see only the shadow of reality. They uses trendline, patterns ,ta to forecast the complexity of non linear market. they tried to backtest using as much years of historical data to find the repeatable pattern of what they called an "edge", however they never realize that the data was already bias and what they called an "edge" is just an illusion...

to find the hidden true order behind the complexity, trader must embrace the uncertainty and define at least some constant in this dynamic world..the reality is not the straight line but the curves evolve at some contract points... one they understand this, they will put big smile under their face
MXN is breaking down in all pairs. KAMA entered EUR MXN (static portfolio) and EDTT in USD MXN (dynamic portfolio).

Attached Image (click to enlarge)
Click to Enlarge Name: RGZckxQ.png Size: 338 KB
Cut short your losses. Let your profits run on. David Ricardo (1772-1823)
Attached Image


6D in time tunnel
Quoting rapidscalper
{quote} Traders are like the prisoners in Plato's cave , they see only parts of the world that are apparent as they see only the shadow of reality. They uses trendline, patterns ,ta to forecast the complexity of non linear market. they tried to backtest using as much years of historical data to find the repeatable pattern of what they called an "edge", however they never realize that the data was already bias and what they called an "edge" is just an illusion... to find the hidden true order behind the complexity, trader must embrace the uncertainty...
The main problem amateur traders face is trying to build too rigid set of rules to trade the market. The market is dynamic and no robotic system can trade it effectively. I don't mean automation is not profitable, but I have tested thousands of systems and my conclusion is that they all work hot and cold. The work 6 months and break even or loose the next. The only way to truly outperform consistently Is to get the general principles and then think, think. Yeah you should have a method of operation but it should be followed with discretion. There are no formulas that work, keep testing and wasting your time.
When you see it, BET big.
Take a break from FF C.
Stop looking for better entry point and start looking for where to take your profits. When I went through countless of my trades over 5 years I realized that if I held unto my winners longer and risk loosing a small profit for a bigger one, I would have come out way better. Instead of 45% win with risk to return of 1.5x, I should have went for 30% win with risk to return of 5x. I often sold a position and bought it back at a higher price, had I held on I would have made killings. If it profits you I m glad my advice , HOLD ON TO THOSE WINNERS.
When you see it, BET big.
0 traders viewing now