Attachments: Phoenix 2007 (new thread)

Phoenix 2007 (new thread)

#501
Nov 3, 2006 12:57pm Nov 3, 2006 12:57pm

blewsky
| Joined Dec 2005 | Status: Trader | 19 Posts

Hello,

I used the following settings:

SMAPeriod 2
SMA2Bars 18
Percent .0032
TakeProfit 42
StopLoss 84
TrailingStop 0
OSMAFast 5
OSMASlow 22
OSMASignal 2
EnvelopePeriod 2
Fast_Period 25
Slow_Period 15
DVBuySell .0029
DVStayOut .0024
PrefSettings false

USD/JPY 3 trades this past week 3 winners

I hope this helps.

Scott

#502
Nov 3, 2006 1:34pm Nov 3, 2006 1:34pm

Craig
Joined Feb 2006 | Status: Blah blah blah | 1,410 Posts

First of all congrats on the MQL contest, nice going. Looking though the trade history it seems that the EA has an amazing win rate, but the losers are 3x bigger that the winners. Two things spring to mind reading the code...

1. Why not calculate lot size on risk & stop loss? this way max risk is actually max risk & lot size increases on account size.
2. Have you considered a more refined MM approach (e.g. break evens etc)?

The breaking of a wave cannot explain the whole sea.

#503
Nov 3, 2006 4:06pm Nov 3, 2006 4:06pm

alamanjani
| Joined Oct 2006 | Status: Trader | 72 Posts

I have one quick beginner question, how do I delete past trades from 'Account History' tab?

I think I read once somewhere, but don't know where and program Help didn't help me.

Thank you in advance

#504
Nov 3, 2006 4:50pm Nov 3, 2006 4:50pm

tradeigel
| Joined Jun 2006 | Status: Trader and Programmer | 393 Posts

Craig:
Please have a look at the new Phoenix - P5! there a breakeven-option, I think.

alamanjani:
I think you can't delete anything. However, you could use the option (sth. linke "customize period") when you use a right-click in acc. hist.

Quoting alamanjani

Disliked

I have one quick beginner question, how do I delete past trades from 'Account History' tab?

I think I read once somewhere, but don't know where and program Help didn't help me.

Thank you in advance

Ignored

#505
Nov 3, 2006 5:04pm Nov 3, 2006 5:04pm

athalon7
| Joined Oct 2006 | Status: I Hedge | 206 Posts

Way to go Hendrick! We know you worked hard on this version. I hope you win the contest. 40g's is a lot of money. To the test lab!

"Its not where your at, but where your coming from"

#506
Nov 3, 2006 5:26pm Nov 3, 2006 5:26pm

alamanjani
| Joined Oct 2006 | Status: Trader | 72 Posts

Quoting tradeigel

Disliked

I think you can't delete anything. However, you could use the option (sth. linke "customize period") when you use a right-click in acc. hist.

Ignored

Thank you!

#507
Edited 5:16pm Nov 4, 2006 3:00am | Edited 5:16pm

quarana
| Joined Nov 2006 | Status: Trader | 1 Post

Hello!

From some time I'm interested in forex market from as a mathematician.
Phoenix seems to be a very promising approximation of the market behaviour.
In fact I'm quite surprised by the fact that the set of relatively simple indicators
implemented proved to be very successful.

I played for a while with a metatrader platform and it`s simulator, which seems
to be far from perfect in it`s treating of time series. So in this situation forward
test are more accurate.

I currently work on custom tick-based simulator and expert advisor taking advantage of higher
frequency signals to predict longer term trends using wavelet mathematics.

And here comes my question.
Many forward test on this list are conducted on demo platforms. This would be fine if the
demo feed matches perfectly life data feed in its characteristics. For example: demo data feed can
change more frequently and oscillate around slower true live feed. In this case indicator`s parameters tuned
on faster data can function less effectively on slower data feed even when working with minute bars.
For example data displayed on [link removed - forum rule #2] website changes slower than on NEXTT demo metatrader data feed.

Hendrick, could you please tell us, whether data on NEXTT demo feed matches perfectly true live feed?

#508
Nov 4, 2006 3:47am Nov 4, 2006 3:47am

Craig
Joined Feb 2006 | Status: Blah blah blah | 1,410 Posts

Tradeigel, so there is, my apologies.

The breaking of a wave cannot explain the whole sea.

#509
Nov 4, 2006 4:45am Nov 4, 2006 4:45am

Benna
| Joined Jul 2006 | Status: Trader | 34 Posts

Hendrick

Africa rejoices!

We concratulate you on your no 1 position!

Ben
Cape Town

#510
Nov 4, 2006 5:27am Nov 4, 2006 5:27am

Benna
| Joined Jul 2006 | Status: Trader | 34 Posts

Hendrick

Thank you very much for your hard work with Phoenix 5. I am sure this one will be the biggest surprise to us all!
We are thankful that you allow us to help you. Just tell us what to do. It is no use for everyone to test the same settings. Give us a broad value for the settings and let us find the correct combinations.

I am confused with the settings1 to 5. They are all zero except for setting4 with fast_price=1 and slow_price.

Will you please tell us what settings should be used for settings1 to 5.

Thanks!

Ben

#511
Nov 4, 2006 5:57am Nov 4, 2006 5:57am

Hendrick
| Joined May 2006 | Status: Trader | 407 Posts

Quoting Benna

Disliked

Hendrick

Thank you very much for your hard work with Phoenix 5. I am sure this one will be the biggest surprise to us all!
We are thankful that you allow us to help you. Just tell us what to do. It is no use for everyone to test the same settings. Give us a broad value for the settings and let us find the correct combinations.

I am confused with the settings1 to 5. They are all zero except for setting4 with fast_price=1 and slow_price.

Will you please tell us what settings should be used for settings1 to 5.

Thanks!

Ben

Ignored

Hi Ben,

There will be a new manual for Phoenix (today or tomorrow) and I'm sure you will find all the answers in there.

#512
Edited 5:17pm Nov 4, 2006 5:59am | Edited 5:17pm

Hendrick
| Joined May 2006 | Status: Trader | 407 Posts

Quoting quarana

Disliked

Hello!

From some time I'm interested in forex market from as a mathematician.
Phoenix seems to be a very promising approximation of the market behaviour.
In fact I'm quite surprised by the fact that the set of relatively simple indicators
implemented proved to be very successful.

I played for a while with a metatrader platform and it`s simulator, which seems
to be far from perfect in it`s treating of time series. So in this situation forward
test are more accurate.

I currently work on custom tick-based simulator and expert advisor taking advantage of higher
frequency signals to predict longer term trends using wavelet mathematics.

And here comes my question.
Many forward test on this list are conducted on demo platforms. This would be fine if the
demo feed matches perfectly life data feed in its characteristics. For example: demo data feed can
change more frequently and oscillate around slower true live feed. In this case indicator`s parameters tuned
on faster data can function less effectively on slower data feed even when working with minute bars.
For example data displayed on [link removed - forum rule #2] website changes slower than on NEXTT demo metatrader data feed.

Hendrick, could you please tell us, whether data on NEXTT demo feed matches perfectly true live feed?

Ignored

As far as I know, NEXXT and IBFX don't have (significant) differences between data feed for their live- and demo-accounts.

#513
Nov 4, 2006 6:06am Nov 4, 2006 6:06am

Hendrick
| Joined May 2006 | Status: Trader | 407 Posts

Quoting Joules360

Disliked

Hendrick:

Congrats on 2006 meta trader championship!

The contest EA has been running 24/7 so I guess its handling news trading fairly well.

Should we shut down on friday with v5 or will Phoenix trade well during news as well?

Thanks,
Joules360

Ignored

Hi,

Personally I never close trades on Friday. As far as I known <?xml:namespace prefix = st1 ns = "urn:schemas-microsoft-com:office:smarttags" /><st1:City w:st="on"><st1

lace w:st="on">Phoenix</st1

lace></st1:City> can handle news events very well. Once I've witnessed that <st1:City w:st="on"><st1

lace w:st="on">Phoenix</st1

lace></st1:City> opened and closed a trade during a spike in only 20 seconds. A profit of <st1:metricconverter w:st="on" ProductID="1.700 in">1.700 in</st1:metricconverter> 20 seconds. So, let this baby run 24 hours a day and don't worry about news events!<?xml:namespace prefix = o ns = "urn:schemas-microsoft-com:office:office" /><o

></o

#514
Nov 4, 2006 6:08am Nov 4, 2006 6:08am

Hendrick
| Joined May 2006 | Status: Trader | 407 Posts

Quoting blewsky

Disliked

Hendrick,

Nice job on Phoenix 5. I will be testing for the next week on the setup. However, I am confused as to which setup to use and the values.

Little History:

Last Sunday I started testing 4.2 and was using a settingphoenix2006_10_9.txt file for my values. I set the prefsettings=false on GBP/USD, EUR/USD, USD/CHF and USD/JPY and had a $14,664.02 profit of a $100,000 demo account.

I did print out the pdf file explaining the setup for Phoenix 5 and have a couple of questions.

1. Do we use the prefsetting=true with the NEW VALUES as indicated by you?

2. I found the settingsphoenix2006_10_9.txt file have more detailed values for the INPUTS....

So...what do we do?

If you have a chance can you help us with the setup screens?

Nice to know you and enjoy the day,

Scott
ABQ, NM

Ignored

Hi Scott,

A new manual for Phoenix will be posted today (or tomorrow).

#515
Nov 4, 2006 9:08am Nov 4, 2006 9:08am

Hendrick
| Joined May 2006 | Status: Trader | 407 Posts

Here the new manual for Phoenix 5. Have fun!

Attached File(s)

The Ultimate Guide for Phoenix.pdf 224 KB | 1,277 downloads

#516
Nov 4, 2006 9:48am Nov 4, 2006 9:48am

wackena
| Joined May 2006 | Status: Trader | 366 Posts

I try it again. These are the results for Phoenix 4.2(.03) on North Finance Demo from Sep 24 to Nov 3. Attached is full report.

<TABLE borderColor=#14d501 cellSpacing=2 cellPadding=2 width=320 border=1><TBODY><TR><TD width="20%" height=4>

</TD><TD width="20%" height=4>wins

</TD><TD width="20%" height=4>loses

</TD><TD width="20%" height=4>ratio:1

</TD><TD width="20%" height=4>P/L

</TD></TR><TR><TD width="20%" height=4>audusd