Double CCI Cross up for testing

Halo all

For the past few weeks i've been looking through the different indicators there are out there for mt4 and found that none really suite my needs. they all have their own system implemented for trading, but they dont fit the way i would want to trade, so i started to look through the programming and ended up learning mt4 as a result.

I started with one system, and would test it and tweak it in Metatrador each time. Each day, the system would evolve a little more, until now, piece by piece, it's something completely different from how it started.

It's derived from 2 theories of mine - price trend through the cross of moving averages, and the strenghth of price direction through use of oscillators. I follow a few of the threads here, and the ones that really catch my interest and are based on these concepts are the Sidus thread, the Bagovino thread, and Dave's trades. If you go and follow their threads, they're methods as distinguished as they are from one another are all great, and they all work a good 80% of the time.

The system i'm implementing uses pieces of all of their methods, but the irony of it all is that it all started from another system very much different those system and has evolved to work upon the same principles - moving averages and oscillators.

I had come across this method one time, and it caught my eye, but i didn't pay much attention to it. I gave it more notice, however, when i saw it posted again and being used in Akuma's Bollinger Squeeze.

50CCI ? 34EMA ? 14CCI ? Zero Line Cross / aka 5034 ? Change in Trend Trade<?xml:namespace prefix = o ns = "urn:schemas-microsoft-com:office:office" /><o></o>

Another setup DrBob created and uses is the 5034. In this case, the 50CCI crosses up (down) above (below) the ZL at around the same time that the price bar crosses and closes up (down) above (below) the 34EMA. In other words, the 50CCI and the 34EMA are in agreement. Be this the time to say that 14CCI must be also above (below) the ZL.<o></o>

You will notice many times that a 5034 setup is the beginning of a new trend; almost every great move begins with a 5034.<o></o>

When do you enter this trade? It triggers when 50CCI crosses and closes above (below) the ZL at the same time 14CCI is already above (below) the ZL and the last price bar is closing above (below) the 34EMA. In that case you want to go long (short).<o></o>

The only and main problem we have with this 5034 setup is that sometimes it reverses back through the ZL a few times in a choppy way. For trying to avoid choppy times DrBob has been filtering his signals with longer timeframes, looking at triangles forming on price bars ? he does not take a signal if price is consolidating inside a triangle, please refer to Figure 2 and Figure 3 ? and avoiding trading when 50CCI is stuck between the 100 lines.<o></o>

Also, many times you will go long (short) on a 5034 and shortly thereafter the 50CCI will cross back down (up) through the ZL and price will close below (above) the 34EMA. In this situation you can reverse your position and go short (long) or at least you can exit your trade to avoid a sure loser.<o></o>

When 5034 works, 50CCI will reach the 100 line pretty quickly. So, if it stalls or takes too long to do so you better protect yourself by quickly moving your stop to break even + 1 or so. Now, once 50CCI breaks that 100 line it usually will move fast in the direction of your trade. Just as an example, again only as an example, we could say that the March 2005?s average peak, after 50CCI cleared the 100 line, was 18 for Russell and Euro. Please refer to Targets and Exits for this subject on how to find stops and targets by keeping statistics."

The system simply started with a cross of Emas, then i switched to this when i came across it, now, it's using an MA cross in addition to the settings above.

A cross of the 0 line on both long and short CCIs, and a cross of both the long and short EMAs.

My theory is as follows - each currency has it's own set of values that work. 14CCI and 50CCI dont work for every currency and might not even be all that accurate, and a 34 ema may be close, but will it work on every currency pair? and if it does, is it close, or is there another ema that is a bit more accurate? And should just an ema be used, or should one go on a cci cross? I've found several successful threads, 3 of which are pretty well known i'd say, and profitable (i tried each of 'em myself), and they work on Ema crosses. Dave's uses the EWO, Sidus uses a 5, 8,18,28, Bagovino's system is similar to sidus except for slightly different values in the EMAs used and the use of an RSI. Oh, and not to mention the famous Vegas Tunnel, also based on moving averages.

blah

just ran into an error in the new order check function and i cant seem to get around it... now all of the sudden the expert is placing 2 positions at once instead of closing the previous position when the cci goes the other like it would before...

blah. im too sleepy. i'll finish it up tomorrow.

I dunno about coding but I need to make a sugestion. I would move your #'s on ma's and indacaters to fibo #' since they hold so much significance in not just forex but the whole universe.

here is a list of these. 1,1,2,3,5,8,13,21,24,34,55,89,144 Your #'s are really close but its the little things that make the biggest differance.

Dave

Wouldn't some code similar to this with a CLOSEBUY or CLOSESELL parameter do the trick. You'd need to massage it some to fit your EA with the parameters to close the buy/sell.

//+------------------------------------------------------------------+
//| Signal Begin(Exit Buy) |
//+------------------------------------------------------------------+



//+------------------------------------------------------------------+
//| Signal End(Exit Buy) |
//+------------------------------------------------------------------+

if (Order == SIGNAL_CLOSEBUY && ((EachTickMode && !TickCheck) || (!EachTickMode && (Bars != BarCount)))) {
OrderClose(OrderTicket(), OrderLots(), Bid, Slippage, MediumSeaGreen);
if (SignalMail) SendMail("[Signal Alert]", "[" + Symbol() + "] " + DoubleToStr(Bid, Digits) + " Close Buy");
if (!EachTickMode) BarCount = Bars;
IsTrade = False;
continue;
}
//Trailing stop
if(TrailingStopMode && TrailingStop > 0) {
if(Bid - OrderOpenPrice() > Point * TrailingStop) {
if(OrderStopLoss() < Bid - Point * TrailingStop) {
OrderModify(OrderTicket(), OrderOpenPrice(), Bid - Point * TrailingStop, OrderTakeProfit(), 0, MediumSeaGreen);
if (!EachTickMode) BarCount = Bars;
continue;
}
}
}
} else {
//Close

Finally.
i had the expert ready a few days ago but what i wanted to do was implement a little something where one could trade on multiple charts at once and not have the expert close the trades, since what it would do before was simply close the last trade, which wasn't a good thing if u were using multiple experts, trading on multiple charts, or trading manually. I was able to do this thanks to the magic number setting.

The system still has no settings as far as to what cci periods and ema periods to use, and i still have yet to implement a few things, but the basics are there. The important part is set - the system trades. The next step is to find the "magic numbers" for each currency. I'll take what i've heard so far into consideration and backtest it out, then if it gives me smooth results i'll take it and let it run for a week strait on a demo account.

Something that i have yet to implement is an account management and lot increase/decrease in relation to the gain/loss of the account. I already know how to do this and will implement this later on, but for now i just want to get the numbers str8 before jumping into the next step. Here is the expert so far for anyone who wants to backtest it. It's designed for MT4 (thus the mt4 extension...)

ok. one big mistake found in my ea.

i found in the tester that it would just keep placing multiple orders, something that definitely needed to be changed, so i went ahead and fixed that error. Now it seems to be trading smoothly. Now i'm just backtesting it to see if i can find some good numbers. Should take me a few days of experimenting here and there, but i should have the numbers before the end of the month.

The best thing to do is to test this thing on a demo acount and over the course of a month see if the demo account grows or not. Once i get the right numbers in for each currency pair, i'll go ahead and add the money management feature. No point in adding it now if the indicator isn't going to work MOST of the time (i'm hoping for a good 75% success rate). I'd rather not add the lot increase yet, what i measure by success is the amount of pips made by the system, not the amount of money it brings in. Example - It's easy to make $10 if you trade 10 lots on a mini account, but it's a bit more difficult to make those $10 if u use only 1 lot worth $1 on that mini account.

Oh, and i went and removed the last version of the indicator from the post since it wasn't right. Here's the newer one.

Interesting concept. I try to find some computer capacity for testing it next week...

finally got the numbers for the euro down.

I also finally took the time to work out all the bugs in the indicator (or most of them anyways...) I was able to implement all of the features that i had mentioned earlier, including the money management lot increase/decrease, which goes up and down in direct proportion with the account.

I also spent all weekend backtesting against past history charts with different combinations of numbers and finally came up with these settings for the euro on this system:

short cci 7, long cci 30, short ema 2, long ema 55, stoploss 90, takeprofit 30.

The indicator will USUALLY close the trade and reverse it in the opposite direction waaaay before the price actually ever hits the stoploss, but that number is in there just to be safe, cuz there were occurences where one CCI would do a ZLCross while the other would linger for awhile, and the price would spike. The system is reliable for that 30 pip take profit, but if you feel pretty sure that the trade is going to keep going in your direction (i.e. an new high on the monthly or weekly charts and this thing crosses downward) then just shut off the expert and let the price ride. That's only if ur that confident tho. in most cases however, this thing pretty much makes constant results as far as backtests on 4 hour charts go. Use your own discretion when trading.

i'm gonna start testing the indicator on other currency pairs, and am going to start work on another (which i may be able to later on add into this one...) EA. basically i'm going to focus a little more on the Ema cross part of this and try to move to smaller time frames in future editions of my indicators. For now tho.... have fun trying this one out. it works on 4 hour charts (i think... use ur own discretion).

I was a follower of woodies cci for months, and I'll be very interested to see what you have done here. Looking forward to your next posts.

oh wow. i just experimented with a stochocastic and an ma cross on 1 hour charts. when i did the ema cross on the 1 hour charts i had pretty good results with fair drawdowns. When i added a standard stochastic of 5,3,3 to the ea as a filter tho out of curiousity, drawdown was reduced to 0. I guess that will be my next project to work on once i'm done with DoubleCCI.

neh... fluke in the numbers... still reprogramming.

As for the double cci cross... right now the price is just "hanging" around in a sideways movement, so i'm not very trusting of any of the signals that it gives at the moment as they would probably just be whipsaws.

When you're unsure - stay out.

Rip1999eternal


Have you done anymore work on this?