A professional expert advisor to trade on currencies with the spread of 2 points on 1-minute intervals. It approves itself especially well on flat markets i\with good volumes. The following specific trading technologies has been realized: working with cross-currencies, automated calculation of lots for a preset exposure, multicurrency trading, automated search for the best modeling period, automated correlation between StopLoss value and coefficient of market confidence, etc. The expert realizes pipping technology, techniques to find shadow StopLoss level...
The expert can show results from 100 to 300% daily with default settings. My company distributed this expert advisor only among our commercial customers before. But I think now that some of its ideas could be useful for many traders. Please note that you should not install this version for real trading. Or at least use it only on the flat market at your own risk.
Contacts:
krivoshey@cyberia.org.ru
ICQ: 249106432
15 comments:12
To post a new comment, please log in or register
Well, the author is "dead", as I can see.
Anyway, try to test this system on next history: EUR/USD, 01-01-2000, m1
You can fix it! It makes 3-4 trades with a small profit; Your EA closes these trades prematurely, then it makes one or two trades with a very high loss totaling 10 times all the profit made from the winning trades. Fix the winning trades to reap the highest profit possible each time from each trade. When you achieve this, your EA would be the best every produced.
Please let me know when you fix it. Email me at renoss@hotmail.con.
What values for each paramteres do you reccomend? For what forex instruments this EA is the best? I don't know mql4 (5 will be soon I think) best. Could you reply:
1/. what the best forex instruments will be for this ea?
The code's definitely entertaining, though there's some obfuscation, if I'm not
mistaken.
I quite like the idea of subsampling the MACD decision to 2^N days... that's quite
clever; I didn't know you could choose variable time periods like that.
Tested this at 90% accuracy, with uncertain results - at first, it generated large
losing trades, but then miraculously made a stream of winning trades to 'even out'
things a bit.
Zonker wrote:
Some entertaining code in this one if you have an evening to kill:).
The essence of its trading logic is to look for reversals between pairs of bars.
eg if in one bar close>open and in the next close<open thats a reversal.
The assumption is that if there are multiple occurances of these in a given time
frame its a sign the market is ranging rather than trending and you can trade on
the swings. It runs various SMA type calculations on the size and number of reversals
to decide whether to enter market and what stoplosses etc it should use.
The way it dynamically searches multiple modelling periods is the highlight in
my opinion. I'm borrowing those ideas for my on code for sure;).
If you intend to use on real money, set variable "AutoLots" to "false"
until you have worked out what it does.. either that or set the "Risk"
variable much lower than the default of 0.5 (0.1 at most).
Quite generous to release to public really, although not much good without manual
if you can't read code..
Some entertaining code in this one if you have an evening to kill:).
The essence of its trading logic is to look for reversals between pairs of bars.
eg if in one bar close>open and in the next close<open thats a reversal.
The assumption is that if there are multiple occurances of these in a given time
frame its a sign the market is ranging rather than trending and you can trade on
the swings. It runs various SMA type calculations on the size and number of reversals
to decide whether to enter market and what stoplosses etc it should use.
The way it dynamically searches multiple modelling periods is the highlight in
my opinion. I'm borrowing those ideas for my on code for sure;).
If you intend to use on real money, set variable "AutoLots" to "false"
until you have worked out what it does.. either that or set the "Risk"
variable much lower than the default of 0.5 (0.1 at most).
Quite generous to release to public really, although not much good without manual
if you can't read code..
Hello Openstorm,
would you answer some questions about this EA, please.
what time frame did you use above to run this back test? have you forward tested
this?
dont you think that 24.93% modeling quality is rather low?
what in your opinon would be the best time frame to run this?
what pair do you suggest?
i mean you say to run this on a more stabile environment.. which pair do you consider
that to be?
i assume that you had a pair in mind whne you designed this EA.
i have read some of the other posts on other forums about your EA and i thought
i would come here to ask you some of these questions.
sorry for all the questions, but inquiring minds want to know =p
Thanks in advance,
Furious_angel
Expert Advisors 
