Menu

Options

Bookmark Thread

First Page First Unread Last Page Last Post

Print Thread

Similar Threads

Cornflower Hourly System 2,025 replies

Special request: Can someone make a Cornflower EA for us? 24 replies

cornflower mq4 request. 1 reply

"Bagovino Method" VS "Cornflower Hourly System" 16 replies

Attachments: Cornflower EA
Exit Attachments

Cornflower EA

  • Post #1
  • Quote
  • First Post: Apr 1, 2009 3:00pm Apr 1, 2009 3:00pm
This is the thread to be used for anything regarding the Cornflower EA - whether it is your trading results of the EA or the testing of various settings of the EA itself.

The Cornflower EA was developed by Postrock, a member here at Forex Factory. The EA is based on the Cornflower Hourly System. The Cornflower Hourly System was brought here to the forum by Lever70, another member here at FF. Whether she is the actual creator of the system, I don’t know. Maybe she will post here to confirm if she is the creator of the system.

Here is where you can get the EA:

http://www.forexfactory.com/showpost...8&postcount=24

For those who want to know the rules of the Cornflower System, click the link below:

http://www.forexfactory.com/showthread.php?t=26205

I look forward to seeing everyone posting your experiences here with the Cornflower EA.


Topgun
"Success is not a fantasy, it's a formula." - MTI
  • Post #2
  • Quote
  • Edited 8:48pm Apr 1, 2009 8:34pm | Edited 8:48pm
I am going to post here a little about how I am currently using the EA. I have 2 demo accounts ... Interbank for the 1 hr chart trades and FXDD for 5 mn chart trades. Here are my settings for each account:

Interbank

Time Frame: 1 hour
Trading Time: 24 hrs a day
Number of currencies I am trading: 10
Bar: 0 (I do not fully understand how the "Bar" number affects a trade, so I just left it on "0".
MA Trigger: 6 EMA (I am testing the 6 ema as the trigger because it always seems to work better for me than the regular CF ma's)
Stop Loss and Take Profit: 30 SL & 60 TP for 9 out of the 10 currencies (the exception is the gbp/jpy of which I have a 50 SL & 100 TP)
trading direction: Long & Short

FXDD

Time Frame: 5 min.
Trading Time: from London Session open at 3am est to U.S. close at 5pm est
Number of currencies I am trading: 10
Bar: 0 (I have the bar at this number for the reason mentioned above).
MA Trigger: 12 EMA (I decided here to use the most commonly used CF ma trigger)
Stop Loss and Take Profit: 15 SL & 30 TP for 9 out of the 10 currencies (the exception is the gbj/jpy of which I have a 20 SL and 40 TP)
trading direction: Long & Short


Keep in mind that I am only testing these settings. I plan to test different stop losses and take profits, and I plan to test different MA triggers, to find the best ema to enter with. Another good thing to test is the "trading direction" setting (the setting is called "positions" on the EA). Right now I have my 2 demo accounts set to take all Long AND Short trades. Actually, I think the smart thing to do is to make the best estimate of the most likely direction the market is going and set the setting for that direction (example, set it "Long" if you think the market is going up for a particular trading day or session).


Topgun
"Success is not a fantasy, it's a formula." - MTI
Hey mate,

I might be asking a stupid question here... but i am not able to get any results for backtesting prior to 2009. It will only spit out resuts from January 2nd 2009, no matter what start date i use.

Any help on this? I am running it with FXDD.

Cheers, and i look forward to contributing to this thread.

MM
"If it bleeds ... we can kill it"
i meant to continue and post in the already opened thread...but this is ok if u so prefer.

I'm uploading again the EA 0.1..it has no changes yet as i'm waiting for good ideas on your side or ideas that you may find in the thread o the strategy itself.

Now, you still dont understand the Bar setting..i will make you undestand it this way..Please EXPLAIN ME what is a "moving average". If you can explain me what a MA is i can let u understand thru your own explanation.
Attached File(s)
File Type: ex4 Cornflower EA v0.1.ex4   4 KB | 1,810 downloads
Postrock,


Being the Author of the EA, maybe you can help with my question above?

Thanks,



Quoting Postrock
i meant to continue and post in the already opened thread...but this is ok if u so prefer.

I'm uploading again the EA 0.1..it has no changes yet as i'm waiting for good ideas on your side or ideas that you may find in the thread o the strategy itself.

Now, you still dont understand the Bar setting..i will make you undestand it this way..Please EXPLAIN ME what is a "moving average". If you can explain me what a MA is i can let u understand thru your own explanation.
"If it bleeds ... we can kill it"
I dont know...the only things that comes to my mind is that my EA has a time filter wich allows the trade to be taken only in between specified hours...

Maybe the broker only keeps that kind of numerical value (on demo) for the last year...but this is just an ignorant guess...

TopGun did you also notice such a problem?
Quoting Postrock
Now, you still dont understand the Bar setting..i will make you undestand it this way..Please EXPLAIN ME what is a "moving average". If you can explain me what a MA is i can let u understand thru your own explanation.
Off the top of my head, I don't think I can give the exact mathematical formula for a moving average, but using a 12 ema for an example, I assume a 12 ema is basically the average price of the previous 12 ema bars. However, I may be wrong. I'm not a mathematical expert regarding forex.

So here is my question: If I input a value of "1", does that mean at the point the previous bar/previous trigger MA closed is going to be the entry point?

And if I input a value of "0" then does that mean when the current bar closes (or maybe just touches?) at the trigger MA, then it will enter?

If this is wrong, then is it possible to give an example of a possible entry price using a bar value of "0" vs. "1" (or any other number you choose)? Some sort of an example regarding this issue would help very much.


Topgun
"Success is not a fantasy, it's a formula." - MTI
Quoting Ari Gold
Hey mate,

I might be asking a stupid question here... but i am not able to get any results for backtesting prior to 2009. It will only spit out resuts from January 2nd 2009, no matter what start date i use.

Any help on this? I am running it with FXDD.

Cheers, and i look forward to contributing to this thread.

MM
Ari,

I'm guessing you are using backtesting software. If that is the case then that is something I have no experience in. The only backtesting I've ever done with any system is visual backtesting ... which is that of looking at the past possible trade entries and exits and then calculating the results by hand (or calculator). So, I have no idea why you can't go back further than January 2nd of this year.

Anyway, I've always heard that backtesting is not reliable as forward testing. Having said that, I hope you are able to figure out how to solve the problem because I'd be interested in the backtest results using various SL's and TP's.


Topgun
"Success is not a fantasy, it's a formula." - MTI
  • Post #9
  • Quote
  • Edited 4:43pm Apr 3, 2009 8:12am | Edited 4:43pm
Correct. So, if you set "Bar" 0 in my EA the last of the 12 bars used by the 12EMA for calculation is the last bar on your chart, the unclosed one.
If you set 1 the last of the 12 bars used by the 12EMA is the last closed one.
And so on backwards.
If you set 30, the last bar used by the 12ema for calculation is the 30'th CLOSED bar (again the unclosed bar is 0).
And so on.

Quoting Topgun_1
using a 12 ema for an example, I assume a 12 ema is basically the average price of the previous 12 ema bars.
Topgun
Now...have you found any good ideas to be put in the EA?
  • Post #10
  • Quote
  • Edited 6:48pm Apr 3, 2009 2:11pm | Edited 6:48pm
Postrock, thanks for clearing up my confusion with the bar value thing. I now understand it.

My next question is regarding the trigger MA. As you already know, the Cornflower system has 4 different ema's: 8, 12, 24, and 72. My question is -- What if I wanted to use a completely different ema for the trigger, such as a different ema from those 4 original cornflower ema's? For example, if I wanted to use the 5 ema as the trigger, could that be done, even though it is not one of the original Cornflower ema's?

In regards to your question of anything else to put into the EA, well so far I can't think of anything else. Once I test it for a while I may come up with an idea or two. However, for right now it seems to be working well (from what I can tell from the short time I've been testing it). I plan to report my trading results for the trades of this week. I'll do this after the close of the market today or sometime during this weekend.


Topgun
"Success is not a fantasy, it's a formula." - MTI
Quoting Ari Gold
Postrock,
Being the Author of the EA, maybe you can help with my question above?
Thanks,

I finally understood what you meant for that problem..its probably just my hour filter wich is badly written.. at the moment it will make problems if you start hour is "bigger" than the end hour.

I will have to rewrite that little segment of code when i have time, no big deal really. Thanks for the input..
As promised, here are the results of my Cornflower EA trades for last week.

1 hr TF Cornflower trades

Pip Total = +870
Won = 46
Loss = 69
Win Percentage = 40%
(There are 5 current trades that were still in progress when the market closed Friday. The eventual results of those trades will obviously affect the above results I posted.)

5 min TF Cornflower trades

Pip Total = -1310
Won = 83
Loss = 249
Win Percentage = 25%
(There are 4 current trades that were still in progress when the market closed Friday. The eventual results of those trades will obviously affect the above results I posted.)


Comments: All my trades have had a 2:1 risk/reward ratio. My ema trigger for the 1 hr trades is different from the ema trigger for the 5 min trades. However, I don’t think that is the reason the 5 min trades had so many losses. I believe it is due more to the fact I was using a 30 pip take profit target for the 5 min trades (except for the G/J, of which I was using a 40 pip target). A 30 pip target is really too large of a target for such a small timeframe as the 5 min timeframe. I knew that from the beginning, but I still wanted to test it anyway. I am now going to test a different SL and TP for the 5 min trades. For at least the next week it will be a SL of 12 pps and a TP of 25 pips (keeping it at least a 2:1 R/R). I still think 25 pips is too large of a target, but I will test it to see what happens. Gbp/jpy will be the only currency I will be using a different SL and TP. It will have a 17 SL and 35 TP.

This week I am also going to begin testing trades on Alpari using a 15 SL and 20 TP (using a 5 min timeframe). If the 20 TP proves to still be too large a target, then the following week I will set the TP at 15.

I really would be interested to see anyone else’s results of the Cornflower EA.


Topgun
"Success is not a fantasy, it's a formula." - MTI
Hi,
you have very intresting results.
I just made a backtest (1H) for march with the same settings as you and did not have such good results (besides eur/jpy that was 350 pips pos ,the others where about b/e together.
maybe you can make a backtest on your platform (interbank) on the same period you tested , with the same settings , to see the results?
(F6, settings, in use date you put the dates, in expert properties your inputs, then start [move the visual mode button to the right to make it quick], then you have results and graph and report to see the results, and in symbol you can chose the currencies)


Quoting Topgun_1
As promised, here are the results of my Cornflower EA trades for last week.

1 hr TF Cornflower trades

Pip Total = +870
Won = 46
Loss = 69
Win Percentage = 40%
[b][size=3][font=Times New Roman](There are 5 current trades that were still in progress when the market closed Friday....
Excellent work. Will follow closely.

Regards,
MM

Quoting Topgun_1
As promised, here are the results of my Cornflower EA trades for last week.

1 hr TF Cornflower trades

Pip Total = +870
Won = 46
Loss = 69
Win Percentage = 40%
[b][size=3][font=Times New Roman](There are 5 current trades that were still in progress when the market closed Friday....
"If it bleeds ... we can kill it"
Not to interfere, as i'm glad that you test the EA in every way u see fit...but 5 min timeframe...i dont know, it doesnt sound very viable i think.

Also.. Topgun, i may have missed this info, but i didnt understand what TP and SL settings were u using on the hourly timeframe. The results look promising.

btw. until i havent rewritten that segment of code please always have your "Start_at" setting SMALLER than your "Stop_at" setting.
  • Post #16
  • Quote
  • Edited 9:55pm Apr 7, 2009 9:07pm | Edited 9:55pm
Quoting Topgun_1
As promised, here are the results of my Cornflower EA trades for last week.

1 hr TF Cornflower trades

Pip Total = +870
Won = 46
Loss = 69
Win Percentage = 40%
[size=3][font=Times New Roman](There are 5 current trades that were still in progress when the market closed Friday....
I'm sorry for being a little late in updating last week's final results. As I mentioned, there were 5 trades on the 1 hr account that were still open at last Friday's market close, and there were 4 open trades on the 5 min account. Those trades completed on Sunday. Here are the final results of last week:

1 hr TF Cornflower Trades

Pip Total = +1080
Won = 50
Loss = 70
Win Percentage = 41.66%

5 min TF Cornflower Trades

Pip Total = -1190
Won = 87
Loss = 249
Win Percentage = 25.90%

In the future, I will wait until ALL trades close before I post any results. This will solve the problem of me having to go back and re-calculate the results, as I had to do here.

EDIT: I forgot to note that last weeks trades were not that of a full week of trading. I was away from my computer several days and did not return home until last Tuesday, 3/31. So, it wasn't until 3/31/09 at 1:18 pm est that the first trade triggered for the week. I just want to make this clear.

Topgun
"Success is not a fantasy, it's a formula." - MTI
Quoting vanupips
maybe you can make a backtest on your platform (interbank) on the same period you tested , with the same settings , to see the results?
(F6, settings, in use date you put the dates, in expert properties your inputs, then start [move the visual mode button to the right to make it quick], then you have results and graph and report to see the results, and in symbol you can chose the currencies)
Vanupips, thanks so much for the info on how to backtest. I never knew I could backtest by using the MT4 platform! I've always done it the hard way - visually and manually. Now it will be MUCH easier.

I'm in the process now of backtesting 5 min trades. I will backtest several SL's and TP's using the 5 min charts.


Topgun
"Success is not a fantasy, it's a formula." - MTI
Quoting Postrock
Not to interfere, as i'm glad that you test the EA in every way u see fit...but 5 min timeframe...i dont know, it doesnt sound very viable i think.

Also.. Topgun, i may have missed this info, but i didnt understand what TP and SL settings were u using on the hourly timeframe. The results look promising.

btw. until i havent rewritten that segment of code please always have your "Start_at" setting SMALLER than your "Stop_at" setting.
Postrock, you may be correct that the 5 min timeframe is not the best one to use. So far I've backtested several SL's & TP's with the 5 min timeframe and none of them have yet to show a profit. I'm going to test a few more scenarios with the 5 mn TF, but I'm not feeling too confident it will turn out any better. However, I want to point out that I am only testing the last 3 weeks of March. I tried to backtest the entire month, but for some reason it will only let me backtest from 3/10/09 - 3/31/09. Strange.

As for my SL and TP settings for the Hourly timeframe, I am using a SL of 30 pips and a TP of 60 pips for the following currencies: E/U, G/U, U/CHF, U/J, U/CAD, A/U, E/J, E/GBP, E/CHF. For the GBP/JPY, I am using a SL of 50 pips and TP of 100 pips.


Topgun
"Success is not a fantasy, it's a formula." - MTI
Quoting vanupips
.....
(F6, settings, in use date you put the dates, in expert properties your inputs, then start [move the visual mode button to the right to make it quick], then you have results and graph and report to see the results, and in symbol you can chose the currencies)
Hi all,

Can someone advise me, how come when i press F6 and nothing popup or anything happen on my Mt4? Please advise.
Quoting fxykl
Hi all,

Can someone advise me, how come when i press F6 and nothing popup or anything happen on my Mt4? Please advise.
Try this:

1. Go to the "View" tab at the top of the (MT4) screen

2. When the box drops down, click "Strategy Tester". The Strategy Tester is what you do your backtesting with.


Topgun
"Success is not a fantasy, it's a formula." - MTI
0 traders viewing now