Profit Generator EA - page 12
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
ok, but as a newbie I'll wait for someone more experienced to give high and low limits to different parameters.
I'm currently testing 2.4 on presets from Holyguy7 and from 1:00 AM I only made 3 trades (1:10 sell usdcad on daily BAD; 7:29 and 14:29 gbpusd on H1 first 28 pips and second actually loser). It seems to be normal, anyone have same results?
Also, to allow multiple timeframes tests, I changed three lines on 2.6 version:
1 -
extern int ID;
becomes
extern int ID_BASE=100000; to give a number to PG_2.6 (here 1)
2 - I added next line just under "int Bar;"
int ID;
3 - I added init() function
int init(){
ID=ID_BASE+Period(); return(0);
}
So on a Daily chart magicnumber will be 101440 and will appear on trade's comment. It will be easier to analyse results for each timeframe and kind of parameters. I didn't test it yet but it should work.
and for a specific chart you will have a fixed ID and not a random number.
Also, maybe we could add a initial takeprofit using superclose function. Just in case of computer crash.
PS: I made other changes that I must delete before posting. tell me if interested.
It could have been great on all JPY if it had comes up a little more.
Maybe we could start by pending orders based on lower timeframe bars (M1?) to look for few pips more (covering spread for example). A kind of trailing start.
Sure P() is better. other changes were about comments but uninteresting.
Who code the superclose function? I am trying to well understand it but to had a initial takeprofit (which should be simple, a value like 2 times takeprofit) maybe the initial coder will do it faster. It's just in case of crash and not the most important while testing. can wait and I will have time to do it when tests are launched.
About tests plan, you seems to be the right person to first define limits to parameters. And after we'll see how many we are and which tests we should do first.
It could have been great on all JPY if it had comes up a little more. Maybe we could start by pending orders based on lower timeframe bars (M1?) to look for few pips more (covering spread for example). A kind of trailing start.
Cool. How about something simple like if the main signal is ready to go long and stochastics on M1 (K<D), then wait till the cross, then get in? ...or did you have something else in mind?
I'm really a newbie (only 3 months that I read a lot and backtest) but I have read only good things about stochs (not enought pratice to have a real advice).
But to keep this EA free of indicators (the initial idea), whe could just use trailing bars on M1? or M5? to be a the best place to trade and avoid trading in the bad way.
But let make the first steps in a real forward testing plan and based on results we'll see better if a system like trailing bars or stochs could have helped in a better entry, no?
Sure P() is better. other changes were about comments but uninteresting.
Who code the superclose function? I am trying to well understand it but to had a initial takeprofit (which should be simple, a value like 2 times takeprofit) maybe the initial coder will do it faster. It's just in case of crash and not the most important while testing. can wait and I will have time to do it when tests are launched.
About tests plan, you seems to be the right person to first define limits to parameters. And after we'll see how many we are and which tests we should do first.Can you restate your question? I don't quite understand it. Sorry.
Cool. How about something simple like if the main signal is ready to go long and stochastics on M1 (K<D), then wait till the cross, then get in? ...or did you have something else in mind?
Nich,
I think using any kind of indicator is a bad idea. The strength of the system is because it depends on price action and adding any derivatives and degrees of freedom will only degrade the performance.
Just my $0.02.
Maji
Test Plan
I don't like to have orders with no SL or no TP and when using superclose(), there is no TP untill Trailing be activated. So I ask why not set an initial takeprofit (before superclose set one) to be secured in case of computercrash. That's why I proposed two times the value of takeprofit variable.
Also to initiate on the test plan, parameters are:
Stoploss: 10 to 30?
Takeprofit: 20 to 100?
TimeFrames: M1 M15 H4 Daily Weekly..
Currencies: 12 from Holyguy7 selection? only those with a little spread (<=5)?
those are just suggestions.
Nich,
I think using any kind of indicator is a bad idea. The strength of the system is because it depends on price action and adding any derivatives and degrees of freedom will only degrade the performance.
Just my $0.02.
MajiKISS-"Keep it simple stupid". I agree with your statement, however JO has a great idea and stochastics on a one minute chart can get you better entry points. For example, we are entering into a long position from 'Price Action'. The final check is to see if (k>d)on M1. If K>D then the order will go through immediately. If not it will wait until it has crossed, and then enter the order. It won't limit the amount of trades; only get better entry points... just my 2 cents
Great! Thanks a lot!