EAmsrc V.6c

New comment
 
 

EAmsrc

Hi,

Maybe the following attachments will help.

EbenB

Files:
eamsrc_v6c.ex4  11 kb
eamsrc_v61.ex4  11 kb
eamsrc_v61.zip  11 kb
 

Hi,

Here's the same system for GBP/USD with others parameters

Nobody interested ?

Michel

 

EAmsrc - another Martingale?

Hi dmbsys,

Thanks for sharing.

I have done some backtesting on your EA - with a modelling quality of 90%, it builds up nicely and then rocketed down.

On the negative side:

This EA is nothing different than most of the progressing systems (the "martingales", the firebird's, the quibster's, etc.) - it blows out one's account after a strong trend in the opposite direction - the EA pumps more and more money into the trade and becomes unsustainable. The EA's strong point ultimately becomes it's downfall....

On the positive side:

Risk maybe reduced by distributing the trades over a few currencies which will hopefully not react the same or show the same behaviour at that specific point in time.

Keep well.

EbenB

 

forward test?

Bonjour Michel,

As with most people in this forum, we all want forward proof that something works. Backtests seem to fail for many reasons. I wish you the best with your EA. Can you answer some of the following questions?

Have you done any forward testing on your EA?

Do you have any live statements available for this EA?

Does it work on any pairs?

Are you planning on trying to sell your EA?

Regards

Putz

 

Hello Putz,

Have you done any forward testing on your EA?

Only 2 weeks

Do you have any live statements available for this EA?

I start in live this monday 21/05.

Does it work on any pairs?

No, with defaults parameters : I trade on EUR/USD and with the last SET file that I post, it's for GBP/USD. Timeframe : H1 for all.

Are you planning on trying to sell your EA?

No.

Just a precision : because the brokers are different, I test my EA only with FXLite and Windsorbrokers.

Thank's and sorry for my english,

Michel

 

Same bar close

dmbsys:
Hi,

Here's the same system for GBP/USD with others parameters

Nobody interested ?

Michel

"good" back testing is due to same bar close within a minute. It cannot be realized in forwarding test. From time to time we were getting high from this kind of EA. It is not real but it do get our heart betting faster when we see the result like this from time to time.

It is might a good EA but have to be forward tested.

 

Best wishes

Hi Michel,

Thanks for your prompt reply.

Good luck with your forward tests. I would strongly recommend starting with a very small live account to make sure it stays profitable for an extended period of time.

Can you divulge how your ea decides when to buy or sell and which indicators are used. I see your code but that is non comprehensible for me.

I will continue monitoring your progress and really hope you have found a stable and sustainable system. I guess that time will tell.

 

Hi Putz,

I use only RSI.

The conditions to open trade are here :

double RSI5c = iRSI(NULL,Periode,FastRSI,0,ShiftValue);

double RSI15c = iRSI(NULL,Periode,SlowRSI,0,ShiftValue);

double RSI5p = iRSI(NULL,Periode,FastRSI,0,ShiftValue+1);

double RSI15p = iRSI(NULL,Periode,SlowRSI,0,ShiftValue+1);

//---- RSIDeltaFilter

if (RSI5c>RSI15c && RSI5p>RSI15p && RSI5c=RSIDiff))

{ myOrderType = 2; }

if (RSI5c<RSI15c && RSI5pUpMax && (RSI15c-RSI5c>=RSIDiff))

{ myOrderType = 1; }

myOrderType 2 = BUY ; myOrderType 1 = SELL

Periode, FastRSI, SlowRSI, RSIDiff, UpMAX & DwMax are extern parameters.

Michel

 

testing the system

Hi to all the followers of Michel's system,

I have had another hard look at Michel's system. I agree fully that one should not rely on backtesting alone (for various reasons) and that forward testing is the ultimate way to go.

BUT: My concern is just that Michel's system is a type of piramiding/ martingale system. These systems have proven to be unsustainable after a year or so. In my humble opinion, please do not forward test with real money, unless you can afford to loose 90% of your capital somewhere in the nearby future.

I also had one look at the graph of Michel's backtesting results and compared it with mine. I am very sure that his modelling quality is much too low to be of any descisive value. Unfortunately I do not have the means to provide better backtesting results i.e. exceeding a modelling percentage of 90.

This is just my opinion - for what it is worth...

EbenB

 

Hi EbenB

Ok to test with a modelling at 99 % but, I have no time too...

Do you want to make this test ?

Michel

New comment
Reason: