Phoenix 2007 (new thread)

Percent              = 0.0032;
EnvelopePeriod       = 2; 

Quoting Trader83

Disliked

Hendrick, I've been studying during last week this problem of when Phoenix should be put on a chart with the help of two terminals and 16 charts. There can be difference if you put EA now and 15 minutes later, for example. However, this difference dissapears in 2 or three trades. After 1 trades, now Phoenix 5 on my computer opens positions exactly as Phoenix 4 Contest in the championship.
This can be explained in the following way. The signals for entering positions appears very often, however if Phoenix has already opened a position, it ignores all signals until this position is closed.
So, you have one position. When you put another Phoenix it opens position, that is different to the first one. However, later, the first position or second one is closed, and if there are not any new signals for entering and the last position is also closed, new positions will be opened at the same time by the same signals.

Ignored

Quoting Hendrick

Disliked

Ok. So the best thing to do is let it run and it will get in sync eventually?<?xml:namespace prefix = o ns = "urn:schemas-microsoft-com:office:office" /><o></o>

Ignored

     Percent              = 0.0032;
      EnvelopePeriod       = 2;
      
      TakeProfit           = 42;
      StopLoss             = 84;
      TrailingStop         = 0;
      SMAPeriod            = 2;
      SMA2Bars             = 18;
      OSMAFast             = 5;
      OSMASlow             = 22;
      OSMASignal           = 2;
      
      Fast_Period          = 25;
      Slow_Period          = 15;
      DVBuySell            = 0.0029;
      DVStayOut            = 0.024; 

     SMAPeriod      = 2;
      SMA2Bars       = 18;
      Percent        = 0.0032;
      TakeProfit     = 42;
      StopLoss       = 84;
      TrailingStop   = 0;
      OSMAFast       = 5;
      OSMASlow       = 22;
      OSMASignal     = 2;
      EnvelopePeriod = 2;
      Fast_Period    = 25;
      Slow_Period    = 15;
      DVBuySell      = 0.0029;
      DVStayOut      = 0.024; 

 if(Symbol() == "USDJPY")
      {
      SMAPeriod      = 7;
      SMA2Bars       = 14;
      Percent        = 0.0032;
      TakeProfit     = 42;
      StopLoss       = 84;
      TrailingStop   = 0;
      OSMAFast       = 5;
      OSMASlow       = 30;
      OSMASignal     = 2;
      EnvelopePeriod = 2;
      Fast_Period    = 25;
      Slow_Period    = 15;
      DVBuySell      = 0.003;
      DVStayOut      = 0.024; 

SMAPeriod      = 7;
      SMA2Bars       = 14
      OSMASlow       = 30;
      DVBuySell      = 0.003; 

Quoting Hendrick

Disliked

No. The Contest version of Phoenix is the classic version.

Ignored

Quoting billworld

Disliked

This point was raised before in this thread, but, I don't recall seeing an answer. Unfortunately, for whatever reason, thread searching isn't working right now so I can't pinpoint the prior post.

But, the question has to do with the Percent values which drive the deviation amount for the envelopes. The point raised before was that nothing below 0.1 could be edited within the Envelope indicator settings (e.g. it reverts to 0.0 for its value) and the evelope that gets created at percentages below 0.1 appear to be so tight as to be meaningless. For example, the default settings for UDFJPY are as follows:

PHP Code

Percent             = 0.0032;
EnvelopePeriod     = 2; 

Bring up the Envelope indicator and insert those settings to see what I'm talking about. The band is so tight that price regularly closes above and below it just about every bar. However, a value of 0.3 yields a "more meaningful" envelope. The docs for the iEnvelope function also show the deviation number apparently imited to the 10th decimal place (at least that's what they have by way of example). I'm wondering if there's just a simple error here which may be causing the 1st signal in actuality to be overlooked by virtue of a decimal precision issue. Or, maybe there's something else we're not aware of going on here.

Hendrick, could you kindly shed some light on this?

Thanks

Bill

Ignored

Quoting billworld

Disliked

In looking through the code and default/optimal settings, one thing has had me confused. Namely, the reason why some pairs default settings have fast_period > slow_period.

For example, the following pairs optimal/default settings have fast_period greater than slow_period:

USDJPY
EURJPY
CHFJPY
AUDUSD

What's interesting about this is that currently JPY pairs appear to be leading in performance on various forward tests posted by members here and of course USDJPY is doing great in the contest. I thought someone had also posted results showing one of the AUD pairs also standing out, but, I can't readily find that post.

Maybe there's something to this? A "slower" fast setting and "faster" slow setting is the answer to the "true" hidden patterns/rhythm?

Bill

Ignored

Quoting billworld

Disliked

Hendrick:

Great job cleaning up and re-organizing the code in v5.x. I like your style!

One very minor thing. In the divergence section that has been streamlined by removing maF1 and maS1, the code is still declaring maF1 and maS1 in the following line

PHP Code

double maF1, maF2, maS1, maS2; 

Thanks for the great enhancements you're making.

Bill

Ignored

Quoting Hendrick

Disliked

Is there someone in this thread that is willing to collect all the proposals that were made to change the code of <st1:city w:st="on"><st1lace w:st="on">Phoenix</st1lace></st1:city>? I've made a list so far of these proposals but I can't find the stupid file anymore. I really haven't the time to search the whole thread again! So if there’s someone out there who want to do this..please!!<o></o>

<o></o>

Ignored

Quoting billworld

Disliked

While I'm new to this thread, I'd be willing to help out with maintaining a consolidated feature request list. I used to own a software company and that was a big part of what I did. However, having just spent several hours over the last couple of days studying this thread, I loathe the thought of going through the entire thing again. If acceptable, I'd be willing to step up to maintain the request list on all future requests.

And, here's my current list

-------------------------------------------------------------
Billworld Phoenix Wish List
[C=Code; F=Forum; S=Strategy, D=Docs; T=Testing; G=General]
-------------------------------------------------------------
Nov 12: [D] enhance docs to explain the strategy (take excerpts from interview on MQL4.com)

Nov 12: [s] create version which scales down as price initially moves against entry, make threshold amount user-defined as well as the number of scales positions (max up to 3); maintain same TP/SL as initial entry

Nov 12: [s] create version which takes every signal and reverses directions on reversing signal (doing so may require taking smaller TPs and tighter SLs)

Nov 12: [D] explain differences between contest settings and optimal/default settings in current versions

Nov 12: [C] create visual indicators to assist with analyzing trades and confirming entries/exits on charts

Nov 12: [T] test the curve-fitting concept to see how well it would have played out in prior history, determine sample period (e.g. 6 month period from mid to end of 2005) and then apply optimal settings from this sample period to backtests performed over the first 6 months of 2006; determine at what point (if any) the sample period settings begin to fail; if patterns exist with respect to the time it takes for new settings to be determined, use this as a guage for timing future optimizations

Nov 12: [T] assign volunteers to do specific tests over specific time periods using specific settings

Nov 12: [F] update first page of thread to represent best "landing" spot with most up-to-date info on the latest developments (e.g. latest builds) of the project
--------------------------------------------------------------------

Ignored

Quoting billworld

Disliked

While I'm new to this thread, I'd be willing to help out with maintaining a consolidated feature request list. I used to own a software company and that was a big part of what I did. However, having just spent several hours over the last couple of days studying this thread, I loathe the thought of going through the entire thing again. If acceptable, I'd be willing to step up to maintain the request list on all future requests.

And, here's my current list

-------------------------------------------------------------
Billworld Phoenix Wish List
[C=Code; F=Forum; S=Strategy, D=Docs; T=Testing; G=General]
-------------------------------------------------------------
Nov 12: [D] enhance docs to explain the strategy (take excerpts from interview on MQL4.com)

Nov 12: [s] create version which scales down as price initially moves against entry, make threshold amount user-defined as well as the number of scales positions (max up to 3); maintain same TP/SL as initial entry

Nov 12: [s] create version which takes every signal and reverses directions on reversing signal (doing so may require taking smaller TPs and tighter SLs)

Nov 12: [D] explain differences between contest settings and optimal/default settings in current versions

Nov 12: [C] create visual indicators to assist with analyzing trades and confirming entries/exits on charts

Nov 12: [T] test the curve-fitting concept to see how well it would have played out in prior history, determine sample period (e.g. 6 month period from mid to end of 2005) and then apply optimal settings from this sample period to backtests performed over the first 6 months of 2006; determine at what point (if any) the sample period settings begin to fail; if patterns exist with respect to the time it takes for new settings to be determined, use this as a guage for timing future optimizations

Nov 12: [T] assign volunteers to do specific tests over specific time periods using specific settings

Nov 12: [F] update first page of thread to represent best "landing" spot with most up-to-date info on the latest developments (e.g. latest builds) of the project
--------------------------------------------------------------------

Ignored

Quoting wackena

Disliked

Phoenix v4.2 results from Sep 24 Nov 10.

<TABLE borderColor=#06d501 cellSpacing=2 cellPadding=2 width=278 border=1><TBODY><TR><TD width="25%" height=4>



</TD><TD width="23%" height=4>

Wins

</TD><TD width="23%" height=4>

Loses

</TD><TD width="29%" height=4>

Profit

</TD></TR><TR><TD width="25%" height=4>

AUDUSD

</TD><TD width="23%" height=4>

16

</TD><TD width="23%" height=4>

7

</TD><TD width="29%" height=4>

-554.00

</TD></TR><TR><TD width="25%" height=4>

CHFJPY

</TD><TD width="23%" height=4>

7

</TD><TD width="23%" height=4>

1

</TD><TD width="29%" height=4>

855.63

</TD></TR><TR><TD width="25%" height=4>

EURCHF

</TD><TD width="23%" height=4>

1

</TD><TD width="23%" height=4>

2

</TD><TD width="29%" height=4>

-565.38

</TD></TR><TR><TD width="25%" height=4>

EURJPY

</TD><TD width="23%" height=4>

13

</TD><TD width="23%" height=4>

5

</TD><TD width="29%" height=4>

598.72

</TD></TR><TR><TD width="25%" height=4>

EURUSD

</TD><TD width="23%" height=4>

4

</TD><TD width="23%" height=4>

5

</TD><TD width="29%" height=4>

-588.00

</TD></TR><TR><TD width="25%" height=4>

GBPCHF

</TD><TD width="23%" height=4>

4

</TD><TD width="23%" height=4>

4

</TD><TD width="29%" height=4>

-701.63

</TD></TR><TR><TD width="25%" height=4>

GBPJPY

</TD><TD width="23%" height=4>

15

</TD><TD width="23%" height=4>

7

</TD><TD width="29%" height=4>

-148.47

</TD></TR><TR><TD width="25%" height=4>

GBPUSD

</TD><TD width="23%" height=4>

13

</TD><TD width="23%" height=4>

3

</TD><TD width="29%" height=4>

1,050.00

</TD></TR><TR><TD width="25%" height=4>

USDCHF

</TD><TD width="23%" height=4>

18

</TD><TD width="23%" height=4>

4

</TD><TD width="29%" height=4>

612.23

</TD></TR><TR><TD width="25%" height=4>

USDJPY

</TD><TD width="23%" height=4>

17

</TD><TD width="23%" height=4>

4

</TD><TD width="29%" height=4>

1,352.78

</TD></TR><TR><TD width="25%" height=4>



</TD><TD width="23%" height=4>



</TD><TD width="23%" height=4>



</TD><TD width="29%" height=4>



</TD></TR><TR><TD width="25%" height=4>

Totals

</TD><TD width="23%" height=4>

108

</TD><TD width="23%" height=4>

42

</TD><TD width="29%" height=4>

1,911.88

</TD></TR></TBODY></TABLE>

This last week lost some equity. Result files, along with settings .doc are attached.

Wackena

Ignored

Quoting Hendrick

Disliked

Maybe you can do some tests yourself next week. Start <?xml:namespace prefix = st1 ns = "urn:schemas-microsoft-com:office:smarttags" /><st1:City w:st="on"><st1lace w:st="on">Phoenix</st1lace></st1:City> at different times and then study your trades at the end of the week to see if you can find something. Thanks for your contribution. I think this issue of the right start time will become more and more important in the coming weeks.<?xml:namespace prefix = o ns = "urn:schemas-microsoft-com:office:office" /><o></o>

<o></o>

Ignored

Quoting wackena

Disliked

Phoenix v4.2 results from Sep 24 Nov 10.

<TABLE borderColor=#06d501 cellSpacing=2 cellPadding=2 width=278 border=1><TBODY><TR><TD width="25%" height=4>



</TD><TD width="23%" height=4>

Wins

</TD><TD width="23%" height=4>

Loses

</TD><TD width="29%" height=4>

Profit

</TD></TR><TR><TD width="25%" height=4>

AUDUSD

</TD><TD width="23%" height=4>

16

</TD><TD width="23%" height=4>

7

</TD><TD width="29%" height=4>

-554.00

</TD></TR><TR><TD width="25%" height=4>

CHFJPY

</TD><TD width="23%" height=4>

7

</TD><TD width="23%" height=4>

1

</TD><TD width="29%" height=4>

855.63

</TD></TR><TR><TD width="25%" height=4>

EURCHF

</TD><TD width="23%" height=4>

1

</TD><TD width="23%" height=4>

2

</TD><TD width="29%" height=4>

-565.38

</TD></TR><TR><TD width="25%" height=4>

EURJPY

</TD><TD width="23%" height=4>

13

</TD><TD width="23%" height=4>

5

</TD><TD width="29%" height=4>

598.72

</TD></TR><TR><TD width="25%" height=4>

EURUSD

</TD><TD width="23%" height=4>

4

</TD><TD width="23%" height=4>

5

</TD><TD width="29%" height=4>

-588.00

</TD></TR><TR><TD width="25%" height=4>

GBPCHF

</TD><TD width="23%" height=4>

4

</TD><TD width="23%" height=4>

4

</TD><TD width="29%" height=4>

-701.63

</TD></TR><TR><TD width="25%" height=4>

GBPJPY

</TD><TD width="23%" height=4>

15

</TD><TD width="23%" height=4>

7

</TD><TD width="29%" height=4>

-148.47

</TD></TR><TR><TD width="25%" height=4>

GBPUSD

</TD><TD width="23%" height=4>

13

</TD><TD width="23%" height=4>

3

</TD><TD width="29%" height=4>

1,050.00

</TD></TR><TR><TD width="25%" height=4>

USDCHF

</TD><TD width="23%" height=4>

18

</TD><TD width="23%" height=4>

4

</TD><TD width="29%" height=4>

612.23

</TD></TR><TR><TD width="25%" height=4>

USDJPY

</TD><TD width="23%" height=4>

17

</TD><TD width="23%" height=4>

4

</TD><TD width="29%" height=4>

1,352.78

</TD></TR><TR><TD width="25%" height=4>



</TD><TD width="23%" height=4>



</TD><TD width="23%" height=4>



</TD><TD width="29%" height=4>



</TD></TR><TR><TD width="25%" height=4>

Totals

</TD><TD width="23%" height=4>

108

</TD><TD width="23%" height=4>

42

</TD><TD width="29%" height=4>

1,911.88

</TD></TR></TBODY></TABLE>

This last week lost some equity. Result files, along with settings .doc are attached.

Wackena

Ignored