Supernova GBP/JPY Mini Trend Catcher | Page 218 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Forex Trend Catcher EA 721 replies

trend catcher fib ea 5 replies

Early Trend Catcher Part2 46 replies

Early Trend Catcher 31 replies

Trend Catcher 5 replies

Attachments: Supernova GBP/JPY Mini Trend Catcher
Exit Attachments

Supernova GBP/JPY Mini Trend Catcher

  • #4,341
  • Jan 18, 2008 6:27am Jan 18, 2008 6:27am
I'm optimizing the beta 3 right now, after I finish i'll try the settings on beta 4 and if it's ok i'll post them here.

P.S.
I'm optimizing it on 6 month history, is it right for supernova?
  • #4,342
  • Edited 6:56am Jan 18, 2008 6:36am | Edited 6:56am
And one more thing, I suggest to optimize it with balance minimum=8500(my starting $ is 10k) and maximum drawdown = 20%. This way it goes much faster

---
Oh, almost forgot i use these settings for 3% risk
  • #4,343
  • Jan 18, 2008 7:01am Jan 18, 2008 7:01am
Hi Gnat-Matt!

Will wait for Rick's direction on testing.

Thanks.

Carol



Quoting Gnat
Hallo Carol!

Thanks for your offer to join the testing jamboree! At first I guess we will have to test all the functions of the EA to work properly - that may better be coordinated by Rick.
Then we will optimize settings for profitability and that may be a group coordinated effort then. There are basic knowledge we can use from the first stage, but this EA is something else and may lead to deviationg results. Still we have starting points.

See you soon!
Gnat-Matt
  • #4,344
  • Jan 18, 2008 8:30am Jan 18, 2008 8:30am
Quoting Eureka
Hi Gnat-Matt!

Will wait for Rick's direction on testing.

Thanks.

Carol
Hi Carol,

Thank you very much for volunteering!

We need to be as efficient as possible with our testing as their are quite a few variable for this EA. My first suggestion would be that we settle on a basic starting point that we can use as a control. This would mean settling on standard account size and money management parameters that we would not vary during this testing phase. That way we can focus just on the entry and exit points.

The variables in this phase would be the non-money management EA parameters, timeframes, and currency pairs. I am open to any suggestions the group has as to approach. There are many experienced traders in this group whose input will make a big difference in approach.

What I like to do is choose 3 currency pairs - 1 screamer like GBPJPY, 1 reasonably active like EURUSD, and one super stable like EURGBP. I will then test these 3 pairs on various timeframes using the same parameters. This gives me a baseline. From there, I begin to diverge depending on my observations.

These are just suggestions. Let's get some more input and come up with a plan.

Rick
  • #4,345
  • Jan 18, 2008 8:35am Jan 18, 2008 8:35am
Hi Rick,

I think this System/EA is a winner and we need to find the right settings, etc. I think we could make definite headway with controlled testing.

Just let me know. Glad to help.

Carol

Quoting rmyers
Hi Carol,

Thank you very much for volunteering!

We need to be as efficient as possible with our testing as their are quite a few variable for this EA. My first suggestion would be that we settle on a basic starting point that we can use as a control. This would mean settling on standard account size and money management parameters that we would not vary during this testing phase. That way we can focus just on the entry and exit points.

The variables in this phase would be the non-money management EA parameters, timeframes, and currency pairs. I am open to any suggestions the group has as to approach. There are many experienced traders in this group whose input will make a big difference in approach.

What I like to do is choose 3 currency pairs - 1 screamer like GBPJPY, 1 reasonably active like EURUSD, and one super stable like EURGBP. I will then test these 3 pairs on various timeframes using the same parameters. This gives me a baseline. From there, I begin to diverge depending on my observations.

These are just suggestions. Let's get some more input and come up with a plan.

Rick
  • #4,346
  • Jan 18, 2008 8:48am Jan 18, 2008 8:48am
Hi all,

I have demo v1.4 for 3 days on 5min TF (using 7 pairs), it always have entries.

I install v1.7 beta 4 today and run for 3 hours already, but no trade opened. I am using default settings. I feel strange why there are no trade opened?

Also I want to confirm one thing, Entry rules enabled: QQE DMI CCI Lag T3 <-- is that ALL condition agreed to open trade or ANY of the entry rule agree only?

Thanks very much


Carlton
  • #4,347
  • Jan 18, 2008 9:36am Jan 18, 2008 9:36am
Hi Rick

Like your thoughts on testing by sticking to some basics and then branching out after things are observed awhile. Not the fastest, but definitely the most methodical, otherwise we all wander aimlessly.

Have a suggestion/request for change to EA that may help identify specific trades. Rather than running a vast number of MT4 platforms, each with a different flavour of preset file, I'd prefer the ability to run less with more charts with different presets. Some might prefer the other way and I don't mind that.

I like the fact that the comment for each trade now specifies the tf it is related to. To make my suggestion work, so you know which file is related to which active/completed trade, could you read a line from the preset file, say, always that for 'Comment4', and add that text to the tf that appears in the comment field against the trade? Make sense? Sound reasonable/easy to code?
  • #4,348
  • Jan 18, 2008 9:43am Jan 18, 2008 9:43am
Quoting fxtradera
Hi Rick

Like your thoughts on testing by sticking to some basics and then branching out after things are observed awhile. Not the fastest, but definitely the most methodical, otherwise we all wander aimlessly.

Have a suggestion/request for change to EA that may help identify specific trades. Rather than running a vast number of MT4 platforms, each with a different flavour of preset file, I'd prefer the ability to run less with more charts with different presets. Some might prefer the other way and I don't mind that.

I like the fact that the comment for each trade now specifies the tf it is related to. To make my suggestion work, so you know which file is related to which active/completed trade, could you read a line from the preset file, say, always that for 'Comment4', and add that text to the tf that appears in the comment field against the trade? Make sense? Sound reasonable/easy to code?
Hi Tradera,

That is a very good idea and, as luck would have it, easy to code. I am waiting from the regulars to hear if I Beta 4 seems clean to use for the upcoming tests. If all the original plus new functionality is OK, then I will generate a Beta 5 that will implement your idea, plus one the Dee mentioned. There is a screen capture function in MT4 that will allow us to generate gif's of the screen on events like entries and exits. I had completely forgotten it existed, so thanks to Dee for reminding me. I will make that an option, but I really think that it has merit past the testing phase. It will provide a record of transactions that can be useful in analyzing past trades.

Rick
  • #4,349
  • Jan 18, 2008 9:53am Jan 18, 2008 9:53am
After making some optimization wanted to check when it is prefered time to run the ea, so I made another optimization run, and noticed something buggy:
Attached Image (click to enlarge)
Click to Enlarge Name: bug.png Size: 27 KB
  • #4,350
  • Jan 18, 2008 9:57am Jan 18, 2008 9:57am
Quoting Alextr
After making some optimization wanted to check when it is prefered time to run the ea, so I made another optimization run, and noticed something buggy:
Thanks, I'll take a look.

Rick
  • #4,351
  • Jan 18, 2008 10:01am Jan 18, 2008 10:01am
Quoting Alextr
After making some optimization wanted to check when it is prefered time to run the ea, so I made another optimization run, and noticed something buggy:
I looked at the code and I never write that comment or change the inputs directly, so that looks like just an MT4 generated optimizer comment. I don't think it is a bug. It looks like the step you specified in the optimizer is 4.

Rick
  • #4,352
  • Jan 18, 2008 10:15am Jan 18, 2008 10:15am
Yes it was generated by the optimizer and the problem is with optimizer as well, it checked the same settings about 100 times, that is defenetely not normal...

BTW, the step which I defined was 1, and the only optimized parameters were start and end time
  • #4,353
  • Edited 11:02am Jan 18, 2008 10:38am | Edited 11:02am
Anyway I finished optimizing laguerre + qqe entries, check it out(optimized on gbp/jpy last 6 month).

My broker time is gmt+1, you need to consider that if you are going to test this settings.
Attached File(s)
File Type: txt LaguerreEntries.txt   5 KB | 428 downloads
  • #4,354
  • Jan 18, 2008 10:43am Jan 18, 2008 10:43am
Quoting rmyers
Hi Carol,

Thank you very much for volunteering!

We need to be as efficient as possible with our testing as their are quite a few variable for this EA. My first suggestion would be that we settle on a basic starting point that we can use as a control. This would mean settling on standard account size and money management parameters that we would not vary during this testing phase. That way we can focus just on the entry and exit points.

The variables in this phase would be the non-money management EA parameters, timeframes, and currency pairs. I am open to any suggestions the group has as to approach. There are many experienced traders in this group whose input will make a big difference in approach.

What I like to do is choose 3 currency pairs - 1 screamer like GBPJPY, 1 reasonably active like EURUSD, and one super stable like EURGBP. I will then test these 3 pairs on various timeframes using the same parameters. This gives me a baseline. From there, I begin to diverge depending on my observations.

These are just suggestions. Let's get some more input and come up with a plan.

Rick
Hallo Rick!

This is a good procedure we can work on without anyone hanging in thin air. The EA runs on both Mini- and Standard lots. As many will probably use it with Minis I guess we should test it on those. It would also be good to test on something like USD 5000 accounts to keep it closer to reality.

Then we may want to gather different brokers that members of the thread, willing to test, are using to see the differences there.

The currency pairs ... could you exchange the EURUSD for something else maybe? Hehehe... I my mind this pair is not representative for any cast of pairs on Forex. It is a lame duck for days just to overreact in the next instance!

GBPUSD may be too volatile for the medium pair? Well, in fact it is all the same in the end. I would just personally not include the EURUSD with a real money setup.

So far so fuzzy... I am sorry, it took me 30 minutes to create this babbling as neither wife nor children accept that I am working.

Greetings!
Matt
  • #4,355
  • Jan 18, 2008 10:44am Jan 18, 2008 10:44am
Quoting Alextr
Anyway I finished optimizing laguerre + qqe entries, check it out(optimized on gbp/jpy last 6 month).
Thanks for posting, Alex! Will look as the bandwagon of gypsies here will go to bed! (That is my family...) :
  • #4,356
  • Jan 18, 2008 10:47am Jan 18, 2008 10:47am
Quoting fxtradera
Hi Rick

Like your thoughts on testing by sticking to some basics and then branching out after things are observed awhile. Not the fastest, but definitely the most methodical, otherwise we all wander aimlessly.

Have a suggestion/request for change to EA that may help identify specific trades. Rather than running a vast number of MT4 platforms, each with a different flavour of preset file, I'd prefer the ability to run less with more charts with different presets. Some might prefer the other way and I don't mind that.

I like the fact that the comment for each trade now specifies the tf it is related to. To make my suggestion work, so you know which file is related to which active/completed trade, could you read a line from the preset file, say, always that for 'Comment4', and add that text to the tf that appears in the comment field against the trade? Make sense? Sound reasonable/easy to code?
Hi Tradera!

Good point there and also about testing we may get this done in a streamlined fashion!

See you!
Matt
  • #4,357
  • Jan 18, 2008 11:51am Jan 18, 2008 11:51am
Quoting Gnat
Hallo Rick!

This is a good procedure we can work on without anyone hanging in thin air. The EA runs on both Mini- and Standard lots. As many will probably use it with Minis I guess we should test it on those. It would also be good to test on something like USD 5000 accounts to keep it closer to reality.

Then we may want to gather different brokers that members of the thread, willing to test, are using to see the differences there.

The currency pairs ... could you exchange the EURUSD for something else maybe? Hehehe... I my mind this pair is not representative for any cast of pairs on Forex. It is a lame duck for days just to overreact in the next instance!

GBPUSD may be too volatile for the medium pair? Well, in fact it is all the same in the end. I would just personally not include the EURUSD with a real money setup.

So far so fuzzy... I am sorry, it took me 30 minutes to create this babbling as neither wife nor children accept that I am working.

Greetings!
Matt
Hi Matt,

Good ideas about brokers and account type and size. Yes, we can definitely substitute GBPUSD for EURUSD. USDCHF would not be too bad either.

Rick
  • #4,358
  • Jan 18, 2008 12:09pm Jan 18, 2008 12:09pm
Quoting rmyers
Hi Matt,

Good ideas about brokers and account type and size. Yes, we can definitely substitute GBPUSD for EURUSD. USDCHF would not be too bad either.

Rick
Okay Rick!

Thanks for that! USDCHF would maybe be good as people will not have too much experience with that and it will be sort of a surprise!

So far the beta4, without special settings and only on QQEA, made 500 USD in the past hours since you gave it out to us. Very precise and apart from the pairs that are just there on my platform to see the EA on foreign soil, it is making a lot of money.

The different TFs are working good on another platform too. So far found nothing to report. But then the EA is pretty young - hehehe! I would love to test through the weekend forward, but will try some basic backtests with old data. Guess the beta4 will be fine there also.

Ok... let me have your thoughts.
Bye
Matt
  • #4,359
  • Jan 18, 2008 1:51pm Jan 18, 2008 1:51pm
Now after I downloaded the latest version, I see that i shouldn't had optimized the previous, seems like thouse settings don't work on this one...

***
Actually they work, just made something dumb before that
  • #4,360
  • Jan 18, 2008 1:56pm Jan 18, 2008 1:56pm
Quoting Gnat
Okay Rick!

Thanks for that! USDCHF would maybe be good as people will not have too much experience with that and it will be sort of a surprise!

So far the beta4, without special settings and only on QQEA, made 500 USD in the past hours since you gave it out to us. Very precise and apart from the pairs that are just there on my platform to see the EA on foreign soil, it is making a lot of money.

The different TFs are working good on another platform too. So far found nothing to report. But then the EA is pretty young - hehehe! I would love to test through the weekend forward, but will try some basic backtests with old data. Guess the beta4 will be fine there also.

Ok... let me have your thoughts.
Bye
Matt
Hi Matt,

Yes, the USDCHF can definitely surprise! The GBPUSD is probably better for the first round of testing, as you suggest.

I am very glad to hear that beta4 is behaving (and that it made 500USD). If it held together for everyone else, then we are good to go with testing.

Have a nice weekend!

Rick
0 traders viewing now