Phoenix 2007 (new thread) - page 61

 
daraknor:
I went through that code. It isn't very high quality, and the lotsi calculations have 32 if else statements that need to be checked every time. The same numbers could be reached simply by rounding down or dividing by 5, rounding down for lotsi >15 and then multiply by 5. I might be more willing to spend time with it if the comments were translated Also, with so many arrays being overwritten so often, it is really difficult to trace execution and data models. It would take several hours to find out what the variables are used for, and where they are changed. Maybe this information is in the comments?

I think the kind of optimization we need is genetic algorithms that are then adjusted for tolerance values. If someone can document how they arrive at the values, I will follow the process through for a couple different currency pairs. Maybe the process used can be abstracted through code but I think it is arrived at through much backtesting.

I saw genetic algorithm optimization available on an EA previously during 'strategy tester'. The problem with this kind of method is that the optimized values may have been a specific response to one time conditions.

I have been playing with a few ideas for adaptive take profit levels. Trailing stops are used in Phoenix currently as a sort of "creeping SL" to minimize losses over time. I need to read through the logs to see if this extends to profit lock in or not. There are a few other ideas for trailing stop methods I have. I'm making a thread called "EA design: Trailing Stop and Take Profit" in the Suggestions for Trading Systems thread. Maybe we should discuss this, and optimization, under that topic?

The EA I forward earlier is a crapt EA. Not usefull EA. Phoenix is much more better. I just give a sample of creating a flat file to store usefull information (settings).

If the Phoenix can calculate its own setting based on past data for example July 2005-Dec 2005 (6 months), then try to use the setting to run on Jan 2006-April 2006, it will be good. So far Jan 2006-April 2006 is not good with current settings of Phoenix.

We need to be ready before the trend changed.

 
Cijas:
1) gratz cause i just won 17.82 $ today on USDJPY (i admit i changed the SL because i have so little account now that i didn't want to lost money after having done some profit)

2) the option i ask respectfully :

As i have a little account (some not many $ to have a marge nor many $ to support the risk of losses) and my broker don't do micro account, can u add an option so in the mode 3 of the last Phoenix it split the order in only 2 orders.

I mean now it do 0.1 lot + 0.1 lot + 0.1 lot and i would like to have only 0.1 lot + 0.1 lot.

Thanks a lot if u can add this option to the EA.

I believe Mode2 only trades 2 lots at a time.

Stop losses are automatically set in mode 3, so that your wins don't turn into losses. (If you are in profit, the stop loss level keeps getting moved up.) That doesn't mean you shouldn't change the stop loss, but it may give you a bit of comfort to know that winners don't become big losers.

 
 
raffaelegalbiati:
Hi all, and really thank you to Hendrick for this wonderful Phoenix. I am following this thread since last August and learrned a lot about this EA. Now I am using it live on an IBFX account since 2 days. Until now I have 7 trades closed with profit and no loss. Of course I am not able to understand how the EA moves the SL anyway I am always worried about breakeven. I saw that at a certain point Phoenix puts a SL at breakeven but I often think to change the Input forcing the system to put a SL at BE when price is 20 pips from entrance point. What do you think about it. Do I risk to loose big moves or it could be a good repair from bigger losses?

Thanks a lot for answering me.

Raffaele

If you look at the Journal you will see that Phoenix moves the Stop Loss higher and higher over time so that winning trades don't suddenly turn into big loosing trades. I talk about the strategy a bit more in https://www.mql5.com/en/forum/175481 I didn't contribute to Phoenix, but reading the code it made sense! Something we should have been doing for a long time. Mode3_MoveSL_Trade_2_3 moves the stoploss up to the order price.

 
Cijas:
1) gratz cause i just won 17.82 $ today on USDJPY (i admit i changed the SL because i have so little account now that i didn't want to lost money after having done some profit)

2) the option i ask respectfully :

As i have a little account (some not many $ to have a marge nor many $ to support the risk of losses) and my broker don't do micro account, can u add an option so in the mode 3 of the last Phoenix it split the order in only 2 orders.

I mean now it do 0.1 lot + 0.1 lot + 0.1 lot and i would like to have only 0.1 lot + 0.1 lot.

Thanks a lot if u can add this option to the EA.

Hi Cijas,

I think this is a justified request. You're very honest about the size of your account and I respect that. So I will make an extra option for you (and others) in the next version of Phoenix. Don't forget to post your results !!

 
fikko:
Hi Hendrick,

The phoenix is Excellent.

However, there are some variables within Phoenix that many be different in the future:

What do you think? If the setting is not much different from month to month, then we do not need this.

Thanks

Hi Fikko!

Very interesting this EA! But I don't think we need this for Phoenix. Looking at the results of the Contest version of Phoenix I think we can use the settings we have now until we get consecutive losses.

Please keep us posted with your great ideas! I really appreciate your contribution to this thread.

 

My modest contribution :

in mode 3, you have 3 trade opened at the same time

and sometimes first TP is reached and price reverse....

A sort of trailing stop :

If first TP reached, SL of 2nd and 3 trade will be moved to the first TP

if second tp reached SL of third trade will be moved to second TP

Perhaps its stupid....

 
bertbin:
My modest contribution :

in mode 3, you have 3 trade opened at the same time

and sometimes first TP is reached and price reverse....

A sort of trailing stop :

If first TP reached, SL of 2nd and 3 trade will be moved to the first TP

if second tp reached SL of third trade will be moved to second TP

Perhaps its stupid....

Not stupid at all! Thanks for your contribution!

 

Historical data

This was my source for historical data:

http://www.alpari-idc.com/en/dc/databank.php

Is it gone? Any alternatives?

 
Hendrick:
This was my source for historical data:

http://www.alpari-idc.com/en/dc/databank.php

Is it gone? Any alternatives?

check out my post #598