Need help with MA crossover system | Forex Factory

Options

Permalinks
Print Thread

Similar Threads

need help: EA for MA crossover Exit 2 replies

need help with a simple MA crossover EA. anyone can help? 7 replies

Need help setting up Crossover alert 0 replies

A "different" crossover system - 600 pips on Friday alone in EUR and GBP 48 replies

Scalping with crossover system 0 replies

Attachments: Need help with MA crossover system
Exit Attachments

Need help with MA crossover system

  • #1
  • First Post: Sep 18, 2006 10:49am Sep 18, 2006 10:49am
Ok i need someones's help. I am trying to develop a programme. There is the 7 sma, 14 sma and 21 sma. Then there is PAR standard default parameters.


Ok when the 7 sma crosses the 21 sma and the PAR is supporting, then an entry is generated. Exit entry is when the 7 sma crosses the 21 sma on the opposite direction. So if the 7 sma crosses above the 21 sma and the PAR is below the bar thats crossing then long, and then when it crosses below the 14 sma an exit is performed, and vice versa. If you know what i mean. A very simple. Almost the same as Mudbuddha but not ema but sma and no need for 15 minute chart PAR confirmation. I use 1 hr chart and the pAR confirmation also from the 1hr chart.
I want an automated system that i can use on fxcm account. Please advice me, if there was a thread about that please let me know, if not tell me where i can get info.

Thank you
  • #2
  • Sep 18, 2006 11:17am Sep 18, 2006 11:17am
It is a great mystrey to me why one SMA crossing another should mean anything. I really would like to know.



Quoting Thebigmo
Ok i need someones's help. I am trying to develop a programme. There is the 7 sma, 14 sma and 21 sma. Then there is PAR standard default parameters.


Ok when the 7 sma crosses the 21 sma and the PAR is supporting, then an entry is generated. Exit entry is when the 7 sma crosses the 21 sma on the opposite direction. So if the 7 sma crosses above the 21 sma and the PAR is below the bar thats crossing then long, and then when it crosses below the 14 sma an exit is performed, and vice versa. If you know what i mean. A very simple. Almost the same as Mudbuddha but not ema but sma and no need for 15 minute chart PAR confirmation. I use 1 hr chart and the pAR confirmation also from the 1hr chart.
I want an automated system that i can use on fxcm account. Please advice me, if there was a thread about that please let me know, if not tell me where i can get info.

Thank you
  • #3
  • Sep 18, 2006 12:06pm Sep 18, 2006 12:06pm
Well go to the 4hr chart and draw the sm lines and pAR and see what would happen if you followed it. You don't get huge pips all the time but your losses are minimal because the 14 sm USUALLY swims between the two lines. So if the trade turns against you the 14 sm will be right on your face pretty quickly. If there is a trend then the trend will keep going.


Why should it mean anything?

Good question, i guess because people use these ma to trade.

So they say, a self fulfulling prophecy. Like fibonacci.
  • #4
  • Sep 18, 2006 12:25pm Sep 18, 2006 12:25pm
I suppose that is as good an answer as any. When I first started trading, I used a combination of 10/30 EMA. Basically, I bought and sold when the 10 crossed the 30. What happened was, I was trading really small and just getting out very quickly and I must have had dozens of small losses. Then one evening, I was sitting in front of the TV with my girl friend when I noticed that buying and selling when the cross-over happend would have been a winner every time. I was looking at maye a month's worth of data.

I almost fell off my seat when I noticed that. Yours truly promptly bought and sold every damn cross-over for a about two weeks successfully. Then I started losing a lot. What happened, of course, was that I was mislead by a trending period and when the trend ended and the markets started to range (as they almost always do), I was taken to the cleaners.

Then I started thinking about what moving averages actually show you. The answer is: EXACTLY NOTHING. If you want to follow the trend (when there are trends), you can just enter when you see the price going up or down because all the MAs do is react to that kind of move. And they get you in too late. So why sit there and watch the price going up and then enter when the move is almost over (which is what will happen maybe 80% of the time)?

Trust me, you are going to lose money using MAs and abandon them (rightly, in my opinion). So you might as well look for something better before wasting your money and time on MA cross overs. What I like to do is trade against cross-overs.

Quoting Thebigmo
Well go to the 4hr chart and draw the sm lines and pAR and see what would happen if you followed it. You don't get huge pips all the time but your losses are minimal because the 14 sm USUALLY swims between the two lines. So if the trade turns against you the 14 sm will be right on your face pretty quickly. If there is a trend then the trend will keep going.


Why should it mean anything?

Good question, i guess because people use these ma to trade.

So they say, a self fulfulling prophecy. Like fibonacci.
  • #5
  • Sep 18, 2006 1:28pm Sep 18, 2006 1:28pm
In fairness, there is no better way to trade than any other way. Period.

Moving averages merely give visual aid to indentify trend and aid in entries as well as exits.

Some traders can see patterns and entries solely on price movement. Some cannot and will not ever be able to. Does that make them less successful? No, not by any means. Crossover systems are VERY popular with stock traders...have always been. They're popular because they work.

If they fail, who's really at fault? The system or the trader for his interpretation of the system?

As long as there are republicans and democrats, liberals and conservatists, skinny people and fat people, there will ALWAYS be a need for systems that view the market in different aspects based on how that individual processes and interprets inputs.

An MA crossover system is still, by far, the best method for new traders to gain an understanding of price (market) movement through observation while using an easy to understand system to gain confidence.
Capital Preservation is key to long term wealth accumulation
  • #6
  • Sep 18, 2006 1:36pm Sep 18, 2006 1:36pm
Quoting jamie12
Trust me, you are going to lose money using MAs and abandon them (rightly, in my opinion). So you might as well look for something better before wasting your money and time on MA cross overs. What I like to do is trade against cross-overs.
Dude, back off! Just because you unsucsessfully tested one of literally hundreds of possible MA crossover systems doesn't mean they can't work. Some of the best systems in existence today are relativley simple concepts. While they are not highly accurate, MA systems have the advantage of never missing a move and some people can handle taking the losses while knowing they will participate in every big move.


This whole experiment is a great exercise along the path of developing a system and learning the characteristics of different trading styles. My advice would be to manually backtest your rules over a sufficient period of time and if you are encouraged by what you see, hire a programmer to code the system into a language supported by your platform. If the results are still favorable continue forward testing in real time while paying special attention to those occasions where the system does not perform well. If you can minimize those situations then just about any system has a decent chance of producing positive equity.
  • #7
  • Sep 18, 2006 1:51pm Sep 18, 2006 1:51pm
I will certainly back off if it makes you happy, mate . I suppose he is gonna lose any way so why not MA cross- overs, eh?


Quoting philmcgrew
Dude, back off! Just because you unsucsessfully tested one of literally hundreds of possible MA crossover systems doesn't mean they can't work. Some of the best systems in existence today are relativley simple concepts. While they are not highly accurate, MA systems have the advantage of never missing a move and some people can handle taking the losses while knowing they will participate in every big move.


This whole experiment is a great exercise along the path of developing a system and learning the characteristics of different trading styles. My advice would be to manually backtest your rules over a sufficient period of time and if you are encouraged by what you see, hire a programmer to code the system into a language supported by your platform. If the results are still favorable continue forward testing in real time while paying special attention to those occasions where the system does not perform well. If you can minimize those situations then just about any system has a decent chance of producing positive equity.
  • #8
  • Sep 18, 2006 1:54pm Sep 18, 2006 1:54pm
come on guys this is not going anywhere... Lets play nice... Scott



Quoting jamie12
I will certainly back off if it makes you happy, mate . I suppose he is gonna lose any way so why not MA cross- overs, eh?
  • #9
  • Sep 19, 2006 1:37am Sep 19, 2006 1:37am
This is what i got from those websites:

Hi Mohamed
We can assist you with your request. Let me explain how we operate. We sell our software/services and then you become a client of ours. At that point we can assist with helping you set up your own algorithms or you have access to ours as well. You get full ongoing technical and trading support, online and telephone mentoring, software and trading manuals and basically everything we have to support you as a client and trader. The automation with the charting application requires your parameters to be built in Intelliscript(charting program language), we can assist with this as it is not difficult and our experience will make it easy. Provided there is nothing too significant you are asking of the technical indicators we should be successful with your objective. The automation is linked to FXCM so when you are ready for live trading we will also assist with your trading account opening with them. You are able to demo as long as you would like. The cost to become a client is $1495.00 USD. Our price not too long ago used to be $3000.00 USD but we have been marketing quite heavily and have reduced the price. The only other cost is the monthly datafeed, it is $119.95 per month. The datafeed is required for live data and auto trading so there is no way around it as we pass the cost from our vendor. If you are an active trader we can waive the datafeed charge if 25 lots on a 100K account are traded each month. It is 250 lots on a mini account for the free datafeed. I'm sure you may have other questions or concerns so please let me know and we can continue to make sure this is the right product for you. We've reviewed the information you have sent and everything appears to be pretty easy for Intelliscript.


So it seems pretty heavy financially.
  • #10
  • Sep 19, 2006 1:57am Sep 19, 2006 1:57am
Here is my plan for getting help:

Educate myself with all the wonderful free posts on forex factory. Help others where possible, keep it balanced.

Achieve mild profitability.

Revisit the situation at the next level and reevaluate.

Maintain educational curve.

demo phase is over and now using a small pip value to preserve capital.

Reinvest in useful materials to hit the next level - if it won't take me to the next level flick it off.

--------------
I wouldn't fork out $1495 for forex education, since most of the education you learn from trading.

Qualifier: that's just me, my personality and my current situation.
  • #11
  • Sep 19, 2006 3:08am Sep 19, 2006 3:08am
True, and thats what i heve been doing for the past 4 months. The time i spent is enormous, but i am beginning to feel that i have not been rewarded. I still enter the market shaky. Its very unpredictable, anyways there is now a time factor. I have been unemployed and thus had time to do all this. I maybe working soon and then how will watch the charts. With limited capital its going to be hard.
  • #12
  • Sep 19, 2006 4:15am Sep 19, 2006 4:15am
Quoting jamie12
Then I started thinking about what moving averages actually show you. The answer is: EXACTLY NOTHING. If you want to follow the trend (when there are trends), you can just enter when you see the price going up or down because all the MAs do is react to that kind of move.
Did you ever think that other traders using different time periods and or different MAs were getting signals to buy when your signal said sell (and vice versa) It told them something. Anyway my point is that it can tell you something but the problem is there are so many MAs being used the signals can get tangled up. You should look that guppy MA setup AKA rainbows with about 50 MAs displayed at once. You'll see there are times when the MAs get tangled up and there are times when all 50 have crossed into either bull or bear. When they all agree, I don't think the signal gets contradicted often.
  • #13
  • Sep 19, 2006 4:50am Sep 19, 2006 4:50am
Jamie,

The answer is quite simple.

When a shorter period moving average crosses a longer period moving average, it means that the price action has changed direction.

The reason people use them is because its an easy mechanical way to verify that the price action has changed direction.

It is easy to calculate and it is easy to automate.

The questions still remain:

1. Why did the price change direction?
2. How far is the price going to move in the new direction?

The traders that can answer these questions are the ones that successfully trade MAX systems.

I hope this has answered your question.

James.

Quoting jamie12
It is a great mystrey to me why one SMA crossing another should mean anything. I really would like to know.
Attached File(s)
File Type: doc 15 WAYS TO TRADE MOVING AVERAGES.doc   24 KB | 2,729 downloads
  • #14
  • Sep 19, 2006 11:38am Sep 19, 2006 11:38am
Quoting Thebigmo
This is what i got from those websites:

Hi Mohamed
We can assist you with your request. Let me explain how we operate. We sell our software/services and then you become a client of ours. At that point we can assist with helping you set up your own algorithms or you have access to ours as well. You get full ongoing technical and trading support, online and telephone mentoring, software and trading manuals and basically everything we have to support you as a client and trader. The automation with the charting application requires your parameters to be built in Intelliscript(charting program language), we can assist with this as it is not difficult and our experience will make it easy. Provided there is nothing too significant you are asking of the technical indicators we should be successful with your objective. The automation is linked to FXCM so when you are ready for live trading we will also assist with your trading account opening with them. You are able to demo as long as you would like. The cost to become a client is $1495.00 USD. Our price not too long ago used to be $3000.00 USD but we have been marketing quite heavily and have reduced the price. The only other cost is the monthly datafeed, it is $119.95 per month. The datafeed is required for live data and auto trading so there is no way around it as we pass the cost from our vendor. If you are an active trader we can waive the datafeed charge if 25 lots on a 100K account are traded each month. It is 250 lots on a mini account for the free datafeed. I'm sure you may have other questions or concerns so please let me know and we can continue to make sure this is the right product for you. We've reviewed the information you have sent and everything appears to be pretty easy for Intelliscript.


So it seems pretty heavy financially.
I wouldn't give up so quickly. There are people who can program in various languages (I'm not one of them) for very reasonable rates. You should be able to do this a lot more cheaply than you were quoted.

Phil
  • #15
  • Sep 19, 2006 1:54pm Sep 19, 2006 1:54pm
I like MAs so this is not a bash against any model that uses them, just a bit of commentary. I've used, and am using the MT4 Strategy Tester in ongoing sets with differing setups, be they top/bottom pickers, trend-followers, breakouts, you name it, i've been testing them and enjoying the research while i trade price action on 30m/60m/Daily charts. Here is what i'm finding in general; I have yet to find a "cross" that traded strictly as a cross, makes profits over a year's period. Here is what i mean. Many will trade say a 13/55 cross (this was made popular by DeMark, the use of those two MAs as an entry, but not on cross, but rather their divergence). What many will mistakenly do at this point is take the cross as an entry/exit signal. This can wipe out accounts quickly during those ranging periods as mentioned in the above posts.

Now take those (or your favorite MAs) and use them as indications of price action/divergence and you are on to something worthy of further study. Think of what the 8ma is doing in the V4hr models against the median 55ma? You don't take the cross, but rather look for a hook in the MAs to signal a possible set up. DeMark has a great chapter in his book "The New Science of Technical Analysis" where he mentions that his 13/55 entry is when you see the first day that both MAs have turned in the same direction, that is your entry signal. Then he gives various example on bar charts as to what he means and how to put it into play while trading. Good stuff. Maybe something to consider in your model. Often time if one waits for a "cross" the price is ready to retrace to some point of past suppor or resistance, pause, then possibly continue on in the direction of the cross.

Just having fun rambling.

Cheers to all,
Thom
  • #16
  • Last Post: Sep 22, 2009 11:06pm Sep 22, 2009 11:06pm
I like MA's and find a great deal of wisdom in all of the posts here.
I am currently working on an EA that uses an SMA 5, EMA 1, and LWMA 20, and an RSI 20. I have tried many of the MA crosses and find they are normally too late for maximum gain and in most cases end up with a loss.

My current parameters are a hybrid of many posts from within the Forex Factory. I also set TP at 400 (in back testing it actually made 328 pips in one trade) and SL is set at 50. Average back testing profits are 22.25 and the average losses are 42.85. Obviously it is not ready for running alone but perhaps by sharing this info there may be some ideas expressed to help refine it to reliability.

The indicators used are

MA's
1, Exponential, HLCC/4
5, Simple, Close
20, Linear Weighted, HLCC/4

RSI based on the previous CLOSED candle.
in MQ4 I use iRSI("GBPUSD",PERIOD_M30,20,PRICE_CLOSE,1)
In my actual code I use NULL,0 instead of the GBPUSD,PERIOD_M30 so that I can move it to any chart when I feel it is ready for prime time.
No Magic - just my current choices. I trade the 30 Minute and currently only the GBPUSD as it has the volume desired for my testing purposes.
In back testing it proves only as a probable solution. As we should all know back testing is only for refining logic and should not be used to prove a system profitable. Only live trading can prove the profitablity.

I will post real results later. I only settled on these parameters today after several weeks of testing numerous variations that wiped out two 5,000 USD Demo accounts.

I now strongly feel that for me the solution is going to be multiple MA's with other confirmation indicator(s).

When the MA1 is 5 pips above the MA20 AND MA1 is above MA5 then BUY
When MA1 Crosses MA20 down then close the trade.

Change 'above' to 'below' for the SELL the change 'down' to 'up' for the close.

I then wait for the rules to reset before attempting another trade. When I did not wait for the reset it caused way too many loss trades.

Looking forward to your thoughts and ideas.
Gedws49
0 traders viewing now