Firebird EA - fixed version - | Page 27 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Firebird EA, please test this one, it is one of the greats 1,228 replies

which version of firebird are you using on live account? 12 replies

Firebird V1.0 an amazing EA ! 33 replies

Going live with Firebird 38 replies

My firebird is a roast cornish game hen 3 replies

Attachments: Firebird EA - fixed version -
Exit Attachments

Firebird EA - fixed version -

  • #521
  • Aug 30, 2006 10:03pm Aug 30, 2006 10:03pm
INteresting statistics Accrete. Thanks for sharing. Other reports posted on the weekend showed that 1.0c wasnt all that hot...now its shining so far this week. Hmm. I think 1.0b will be the leader when we report the results on Saturday though...thats just my bet. A lot of the float will become booked pips with the news due tomorrow and Friday.



Quoting accrete
Hi Wizard, thx for the insights. I think in a live account i might go ahead and at least turn off the ea during the actual announcement, and maybe even close out all trades if it is a big day such as NFP. This Friday? Well i'll probably let it just go since it is only a demo and i'm trying to get the feel of it all.

Art1 hope you figured out those losses. I'm not trading that pair, but so far my 1c is the heads-above winner. The only losses i've incurred have been with the 63G-TS in my testing (but to soon to tell anything of real meaning). . .after NY close:

EA demo as of this post at 6PM Pacific Wed IBFX demo all at default settings, opened this last Sunday:

v1-0b2 closed +859 float -797 net = +62
v1-0c1 closed +1165 float -439 net = +726
63G-TS closed +636 float -707 net = -71

======On another note/thought. What if a trend factor of some sort was thrown in the loop? Like take X-kind-of-trade (long or short) if Xtrend (4 hr, daily, weekly??) was X? For instance we know from typical TA that a spike off support in an uptrend will usualy be indicative of a positive-pip-trade? And vise-versa w/downtrend bounce (spike) off resistance. This might be as simple as looking at the relationship of two MAs such as the 13 & 55 (insert your favorite pair/cross here) cross on XtimeFrame? If SlowMA is above FastMA then take spikes off support in direction of SMA/FMA trend?

Cheers,
Thom
  • #522
  • Edited 2:17am Aug 31, 2006 1:26am | Edited 2:17am
Yardie,

Well, I also made lots of pips with version 1.0b but I think the issue is caused by hedging market orders instead of pending orders as I stated in details in my earlier post here: http://www.forexfactory.com/forexfor...&postcount=510
So, I think placing pending orders would - at least partly - be a solution for the increasing float.
This way we would lose approx. 10 pips per hedge order but could get rid of the high float.

Quoting Yardie
I am not having the excessive loss error with Firebird at the moment so I cannot comment much on that. Maybe Wackena or one of the other coders can provide some more assistance.

So far this week I have made over 1200 pips, yet I am in deficit for the week. I suspect that 1.0b doesnt like tight ranging markets. It likes when the market moves up and down. The past day or so the range in the eurusd has been a mere 50 pips..the same with the usdcad. I think things should be shaken up tomorrow with news and Friday with NFP.
  • #523
  • Aug 31, 2006 1:32am Aug 31, 2006 1:32am
1trader2,

Thank you for your reply. I am not a programmer but the routine for "placing a hadge pending order then liquidate the other once one side was activated" is ready and available since straddle EAs use this technique. I think the application of that routine to Firebird could be resolved by a coder.

Quoting 1trader2
This is a good point. Maybe we need to make a variation of this version and compare the two. The new version would close out or liquidate the other order once one side was activated. I think that this would help the float also. Any other thoughts on this? Is there anyone willing to program this change?
  • #524
  • Aug 31, 2006 1:49am Aug 31, 2006 1:49am
I don't see how a hedge can cause a extra float.
Besides the costs for buying lots needed for building the hedge everything should even out AFTER the building of the hedge.
When the hedge is build when the float is -300 pips that float will never increase because the profit and loss from thereon will even out.
But the situation can last a long time. Assume it takes 2 months to return to the right level. That would be 2 months of tied up money.
Say the hedge saves you 500 pips in losses but also keeps you out of 800 pips of gains because it tied up all your funds we still are on negative ground.
The only ways around that are no hedge or try to detect the reversal point.
Drop the hedge (profit) then wait for the TP. Float increases again but you have profits also locked in.

An aditional problem is whipsaws. Like always....
May the pips be with you.
  • #525
  • Aug 31, 2006 2:24am Aug 31, 2006 2:24am
TraderSeven,

You are right the profit and loss from hedge will even out - but only till one of the orders reaches TP. After that the reamaining order increases the float for a while.

Quoting TraderSeven
...
When the hedge is build when the float is -300 pips that float will never increase because the profit and loss from thereon will even out.
...
  • #526
  • Aug 31, 2006 3:33am Aug 31, 2006 3:33am
Quoting dcraig
Benna,


THANKS for responding to my frantic email. I'm anxious to try your suggestion - where is the "home" button? I'm a little embarrased, but I can't find it ...

Thanks again for responding.

Doug
If there is no home button, use the rolling wheel in your mouse, ore use page up and page down buttons
  • #527
  • Aug 31, 2006 8:57am Aug 31, 2006 8:57am
Well since i posted my test numbers at NY open and close here are today's NY open numbers:

EA demo as of this post at 6AMish Pacific Thursday IBFX demo all at default settings, opened this last Sunday:

v1-0b2 closed +958 float -820 net = +138
v1-0c1 closed +1325 float -450 net = +875
63G-TS closed +635 float -683 net = -48

The numbers were actually adjusted while i wrote this...they went into more net profit. The largest being the b and c versions. Anywhoo like we all know, it's fun to track but to little info to call. For myself it will most likely be that i keep testing for several months before jumping live into this, or maybe just a very small expendable duplicate micro-lot account for the "fun" of it. I still must state mater of factly that those 800ish pips in the c version would have been the most stress-less pips in a live account that i've ever made in a month...and these are only after 3 trading days!

One other little test i've had going is the v63G-TS default but on a 15m chart instead of the 1m. It finaly had some trades kick in after about 12hrs being on: closed +135 and -2 float. Very odd to me though is the closed trade log showing +75, +58.69, +7.81, and a negative -6.75?? on the usdcad. That must have been the trailing stop kick in at an odd level? Anyone else showing odd closures on the TS version?
  • #528
  • Aug 31, 2006 9:23am Aug 31, 2006 9:23am
Quoting A1Cruiser
TraderSeven,

You are right the profit and loss from hedge will even out - but only till one of the orders reaches TP. After that the reamaining order increases the float for a while.
I assume the hedge has no TP so it are the original orders that have the TP hit. If my assumption is correct I think dropping the hedge should be done more intelligently.

Say we hit -300 and thats our point we build a hedge.
We get hit by another 1000 pips but the hedge evens that out
After that 1000 pips price goes back in our favour and moves to our TP of 30 pips. The hedge gets dropped. Is that how it works now?
Ideally the hedge should be dropped at that 1000 pip mark. It cashes in 1000 pips. The float will suddenly jump 1000 pips. But in a way it's still hedged by the 1000 pips in your account so in a way it's still hedged. And more importantly it stops a margin call.
On top of that we collect the normal profits.
It's hard/impossible to detect that turning point but I think the aim should be cashing in on the hedge too.
May the pips be with you.
  • #529
  • Aug 31, 2006 9:51am Aug 31, 2006 9:51am
Quoting accrete
Well since i posted my test numbers at NY open and close here are today's NY open numbers:

EA demo as of this post at 6AMish Pacific Thursday IBFX demo all at default settings, opened this last Sunday:

v1-0b2 closed +958 float -820 net = +138
v1-0c1 closed +1325 float -450 net = +875
63G-TS closed +635 float -683 net = -48

The numbers were actually adjusted while i wrote this...they went into more net profit. The largest being the b and c versions. Anywhoo like we all know, it's fun to track but to little info to call. For myself it will most likely be that i keep testing for several months before jumping live into this, or maybe just a very small expendable duplicate micro-lot account for the "fun" of it. I still must state mater of factly that those 800ish pips in the c version would have been the most stress-less pips in a live account that i've ever made in a month...and these are only after 3 trading days!

One other little test i've had going is the v63G-TS default but on a 15m chart instead of the 1m. It finaly had some trades kick in after about 12hrs being on: closed +135 and -2 float. Very odd to me though is the closed trade log showing +75, +58.69, +7.81, and a negative -6.75?? on the usdcad. That must have been the trailing stop kick in at an odd level? Anyone else showing odd closures on the TS version?
I am running forward test on v1oc1 since 29th Aug 2006 with 8 pairs of currency, the current status is nett profit of $10036 including the float of $5484. The initial depost is $41677 ( after stopping from other strategy ). There were 73 trades closed with postive profit and 23 pending trades ( float trades ). Hence in 2.7 days a nett pips of 1003 is gained or 371 pips gain a day.
  • #530
  • Aug 31, 2006 10:01am Aug 31, 2006 10:01am
Another $1k profit is added 10 mins ago.
  • #531
  • Aug 31, 2006 10:14am Aug 31, 2006 10:14am
I also notice that the float is maintained around 8 to 10% of the capital under the normal market condition. But it may go as high as 20% if the spike is against the strategy with a swing of 100pips on some pairs of currency.
  • #532
  • Aug 31, 2006 10:19am Aug 31, 2006 10:19am
my live FB cleaning up again today thanks to chicago PMI, but I'm going to close all trades and shut it down before bernanke at 5.30, and restart next monday.
TRADE FOR SWAP
  • #533
  • Aug 31, 2006 10:42am Aug 31, 2006 10:42am
I am going to let all 8 of my FB demos (various flavors) rip right thru the NFP as usual. That is what demos are for I think.

By the way, my LIVE account on Firebird will be off until nest Tuesday!
Don Life is expensive, but includes a free trip around the sun.
  • #534
  • Aug 31, 2006 12:34pm Aug 31, 2006 12:34pm
I noticed that the ea indicates non trading hours during the us session. What hours are recommended?
  • #535
  • Aug 31, 2006 1:43pm Aug 31, 2006 1:43pm
TraderSeven,

Thank you for your reply, you have some very good points.
If the hedging worked as you wrote it down, then the float would not increase. But I think now the hedge orders work on the same way as the normal orders, that is, using the settings of 30-300-30 and assuming the market goes to one direction for 300 pips, taking only one hedge order into consideration, there will be 30 pips profit and 300 pips loss, the balance is -270 pips. During the way to 300 pips, until we reached the first 30 pips the two orders even out but beyond 30 pips the negative float will increase.

The ideal case would be if the market turned back right after 30 pips and would go back 60 pips from there giving us the possibility to take out the other 30 pips profit. So, a ranging market is advantegeous for this trading system.

Getting back to the example above, in reality, I think (but am not sure) hitting the first 30 pips and taking profit initialises another hedge order and so on and so on. Consequently, when we reach 300 pips we will have 10*30 pips profit (leaving the spread out of the calculation) and 9 pieces of floating losses: at -30; -60; -90; ... -270 and a 300 pips loss. At this case the net balance is zero and the negative float is 1350 when we reach the milestone at 300 pips.
If I was right, a trending market would be disasterous with this model.

Using pending orders instead of market orders when hedging and keeping live only the "right" half of the hedge pair and going along that 300 pips route we could reach 7*30 pips profit and would have 10 pips positive float assuming the gap were 10 pips at each order (300 = 8*10 gap + 7*30 profit + 10 float).

Any advice would be appreciated.


Quoting TraderSeven
I assume the hedge has no TP so it are the original orders that have the TP hit. If my assumption is correct I think dropping the hedge should be done more intelligently.

Say we hit -300 and thats our point we build a hedge.
We get hit by another 1000 pips but the hedge evens that out
After that 1000 pips price goes back in our favour and moves to our TP of 30 pips. The hedge gets dropped. Is that how it works now?
Ideally the hedge should be dropped at that 1000 pip mark. It cashes in 1000 pips. The float will suddenly jump 1000 pips. But in a way it's still hedged by the 1000 pips in your account so in a way it's still hedged. And more importantly it stops a margin call.
On top of that we collect the normal profits.
It's hard/impossible to detect that turning point but I think the aim should be cashing in on the hedge too.
  • #536
  • Aug 31, 2006 4:34pm Aug 31, 2006 4:34pm
Quoting divinefund
I am running forward test on v1oc1 since 29th Aug 2006 with 8 pairs of currency, the current status is nett profit of $10036 including the float of $5484. The initial depost is $41677 ( after stopping from other strategy ). There were 73 trades closed with postive profit and 23 pending trades ( float trades ). Hence in 2.7 days a nett pips of 1003 is gained or 371 pips gain a day.
Hi Divinefund,
Did you use the money management routine and MaxOpenOrders routine ? Did you use the defaut parameters ?
Thanks
  • #537
  • Aug 31, 2006 4:41pm Aug 31, 2006 4:41pm
Just to tell you guys, that my FB 1.0a was beaten by all the negative swap last night (wednesday). I counted about 300 pips loss due to swap, cause the float was big. If ever I go live with this, I am thinking about closing all trades that would lead to negative swaps each wednesday, but I am not sure if this is not going to interfere with the way FB works.

Any advice is welcome.
  • #538
  • Aug 31, 2006 6:42pm Aug 31, 2006 6:42pm
Although this wont affect some Firebird EAs, I wonder if it will affect the hedge version.

Here's the email I received.

Dear Interbank FX Customer,

We are putting into place some company policies that will directly affect your account. As of market closing on Friday Sept. 15, 2006 we will no longer allow any one account to have more than 100 open trades. If you have more than 100 open trades after this date you will not be able to open any new trades until your total open trades falls under the 100 trade limit. Your open trades will NOT be affected. You will simply receive the error message "Too Frequent Requests" if you try to open another trade
Mono
  • #539
  • Edited 7:26pm Aug 31, 2006 7:02pm | Edited 7:26pm
Quoting kingforex
Hi Divinefund,
Did you use the money management routine and MaxOpenOrders routine ? Did you use the defaut parameters ?
Thanks
The MaxOpenOrders = 10 and MM is false until I changed it to true a day ago. I use Risk = 5, hence all my TakeProfit:Stoploss ratio are changed accordingly, that is if my Takeprofit is 20 then stoploss is 100.
  • #540
  • Aug 31, 2006 7:22pm Aug 31, 2006 7:22pm
Thanks for the info on that email. Yes interesting. I have yet to count the open trades, seems to be around 6 or so per traded pair at any given time?

EA demo as of this post at 6PMish Pacific Thursday IBFX demo all at default settings, opened this last Sunday:

v1-0b2 closed +1152 float -757 net = +395
v1-0c1 closed +1620 float -345 net = +1275
63G-TS closed +1044 float -490 net = +554

Not that i can directly compare the test cuz i only began this one 24hrs ago...on a 15m chart instead of the 1m charts above:
63G-TS closed +272 float +47 net = +319

Wow what an interesting day for the testing here. The largest turnaround was the 63G-TS which added 600ish pips. Some of the trades are 50+ pips! I clearly need to study what is going on behind the scenes here. If one wishes to see the trading logs, they can be found here, under the 08-31 date.

Take care. Tomorrow is the big day! I'm leaving all the eas just a purrin', hey it's just virtual dollars anyway : )

Cheers,
Thom
1 trader viewing now