After reading codersguru's lessons me and the expert advisor wizzard got this far with creating code. I'm just trying to learn to make EA's. I would prefer someone who knows what they are doing to do this but I do want to learn too so I'm trying to make a start at it at least.

This is basically to define the things I'd like to see added to the EMA CROSS already created by coder's guru. Man I would love to see HIM take on this upgrade instead of leaving it to a novice like me.

I like the EMA CROSS system perhaps because I've spent alot of time studying moving average dyanamics. It's something I can get my brain around. What you will notice is that I like to give the user the option to engage certain functions or not in a strategy. This allows for many variations in testing different combinations which forms the basis for comparing and re-evaluating results. I find such compare and evaluation reveals what works well together. I believe in giving users the choice.

I get to explain a couple of the parameters here and where I found the ideas...

I've watched a few different systems operate now. I've gotten ideas from them I'd like to see all incorporated into the EMA cross.

The basket profit idea I got from watching the divergence trader EA which opens multiple orders and pools the equity from them to reach a target and then closes all the orders in the pool together. It occurred to me that this might be an effective method to clean up positions which get 'left behind' in the present EMA CROSS- EA. These positions which go unprofitable need to be cleaned up. I don't know if it will work but it's an idea. I hate to see the EMA CROSS method give back so much after it's made such good gains. It's all about damage control before the damage gets big. clean up the small messes as they happen. that's the idea with this function. (optional of course)

I'm wanting other functions which aren't in the present versions of the EMA cross, (or anywhere else I know of) I'd like to see the ability to execute on a custom bar period. I have no idea how to pull this off but I like the concept since I could do that in my spreadsheet and found several advantages to doing it. Why can't it be done in the metatrader? I think just because no one I know of has DONE it yet. I'm told it can be done buy building the candles from one minute data. There is a custom candle indicator already but I don't know if it engages EA's to execute on that custom candle data or not...well I'd like to see this done with the EMA CROSS method because my spreadsheet told me how advantageous it is to do that with moving average cross strategies. Trust me it matters.

I'd like to see the ability to scale stop losses and trailing stop losses. If we can scale into and out of positions then we should be able to scale down stop loss and trailing stop losses to close positions based on conditions we like. It needs fleshing out but the general idea is to make some conditions to get the most out of closing strategies thru creative stop loss use.

I'd like to see the ability to scale lot sizes because presently the EMA Cross places two orders at the same time one buy and one sell. This seems like hedging at 100% and that doesn't seem like having much confidence in the signal cross to me. Again if the lots were scaled so the opposing direction were half the size of the signal direction then that would minimize losses right there making less drawdown to recover in the first place.

I put a close all hotkey option in this so that with two keyboard keys the user could manually close all orders. That could be for damage control AND for discretionary profit taking. It just keeps the discretionary trader able to take immediate action which could be good for some highly intuitive types of people or whatever.

The specified trading hours is not new but I've not seen the ability to go in and out more than once so I made three sessions for engagement during hot trading hours.

I also made a spot for a MACD and a Stochastic to join this EA if desired. I've seen the three work well together in the CAT50fx system which I've seen work in other places too. It's optional if this can be created like I envision it.

The reversal option already exists in the present version of the EMA CROSS. I just made a place for that here even though I don't fully know how it's done.

In fact I don't know how alot of this is done or how much is doable. All I know is what I'd like to see done. I assume that coder's guru doesn't have the same limitations that I do when it comes to what is 'doable'. I can tell you I know I don't have the understanding to do all this right now. But I don't think that should stop me from envisioning or communicating what I can envision. Envisioning is something I CAN do. So this is just doing what I can. I know it's not much. I hope I'll get some help with this (from real programmers) doing what I'm not ready to tackle myself. I'll keep learning but all assistance is appreciated. This is just trying to communicate what I'm envisioning at this point so someone else who codes can see what I'm thinking. As I learn more I'll communicate what I learn how to communicate. I'm expressing more than I did when I first came here. I didn't even know how to start to make an EA. Now I have all these ambitious ideas (ambitious for me anyway).


If I had to prioritize all this I would say the most important is the function to scale lots because I see that as reducing losses which the current version really needs from my testing of it.

Second importance would be the trading session hours that lets the user put the EA on standby during certain times of the day.

the rest of this is of interest because of what might make a flexible EA for strategy testing purposes but those two functions I can see giving immediate reward to the present EA.

I suppose the rest is just to give me something to learn on as I try to learn to make EA's and do coding. It may add something to the outcome and it may just help me learn to program.

the rest of what's there is probably not necessary to do any explaining...it's pretty common

hey wow cool i just learned how to post alot of code and wrap php brackets around it

Thanks for posting an explanation of the commands! I've been really getting into the system as of late , and I've been wanting to translate some of my trading system ideas into EA's for backtesting. I have been having some difficulty finding tutorials or explanations on the Metatrader language. "Reverse engineering" EA's from the top coders ain't easy. Your trading idea is also interesting. Let's hope one of the forum's best coders can help out on this one!

Thanks rainey,

First let me put you on to the tutorial I've been looking at.
http://www.lightpatch.com/forex/mt_yahoo/MQL4_Tutorial/

I know the reverse engineering blues. I'm passionate about this system right now. I don't have a better grasp on anything better. I am open minded this is just my best going concern at the moment. But my latest tests on this aren't bad, see attached. I have managed to alter the code to allow me to use a TP of 8. That's about as far as I have learned how to modify it using the tutorials. I change one EMA from 2 to 1 and it made a huge difference in the profit generated. I don't recommend the confirm order option in this it only decreases the successful outcome.

If you look at this chart you see it makes alot then gives half of it back at the end. I believe that if the lots on the hedged positions were scalable that this amount which is given back would be reduced. I'd like codersguru to make this modification or someone else competent. I'd just like to see that much done with it and see where it leads. All the rest of the stuff is extra credit so to speak.

I'll keep learning as best I can but I really wish someone competent to program would take an interest in this soon. I think that would be best. All I can do is all I can do.

Attached Files

	EMA_CROSSmodv2ky 8TP2L.htm (120.4 KB, 70 views)
	EMA_CROSSmodv2ky.mq4 (10.9 KB, 59 views)

this is along the lines of what I had in mind to scale the stops...

http://www.forex-tsd.com/suggestions...loss-code.html

Aaragorn,
I looked at your back testing results and I noticed that your modeling quality is only 29% or so. I don't mean this in a mean way, but those results mean nothing. Even with a modeling quality of 90% the results really don't correlate with live trading too well. Back testing with a modeling quality of 90% just starts to give a trader an idea of what potentional the EA Might have. If you really want to see what an EA can do you have to forward test it at present. It is possible that in the future a better backtester will come out, but for now it's not all that reliable. Just try forward testing an EA and you will see what I mean.
Aplari's data that can be download to get modeling quality to 90% is indicative data and therefore doesn't accurately represent the true market movement but only is an estimate that may or may not give signals in live trading. That is why it is hard to trust the backtester.

That is all I have for now. I'm not trying to burst your bubble or anything, I just want you to be aware.

Thanks
Huhenyo

Hi Aaragorn,
I've made a few updates, let me know if they are what you want.
In the inputs for the EA you can now tell the EA when to start trading and when to stop trading. There is also the option to specify the % of the lots you want the hedged position to be.

Huhenyo

Attached Files

EMA_CROSSmodv3ky.ex4 (11.8 KB, 39 views)

I understand what you're saying about the modeling quality. I know it's indicative not conclusive but indicative is ok to tell me what to examine in the system.

Would you repost this upgrade in .mq4 format, I'm not sure how to handle it in the format you've used.

Thanks for your kind upgrade, I'll get back as soon as I have some idea of what is working, but i need the right file format to do that.

Aaragorn,
There you go. Sorry, I accidently gave you the wrong file extention. This should work for you now.

Attached Files

EMA_CROSSmodv3ky.mq4 (11.5 KB, 57 views)

Here is the EA I've been working on for awhile now. Try it out and see what you think. I'll post two versions. The 1.1 version only trades with the signal. The 1.2 verions trades with the signal during less volitile times and opposite the signal during the more volitile times with a traling stop only if the movement after the signal is such to warrent it. Have fun. Let me know what you think

Huhenyo

Attached Files

	ConservativeAnt1.1.mq4 (11.0 KB, 45 views)
	ConservativeAnt1.2.mq4 (14.8 KB, 50 views)