Kang Gun Relative Strength trading - the EA's | Page 6 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Trading system using relative strength 2,816 replies

Trading System Using Relative Strength-Part 2 1,499 replies

Updated CCFp and CFP indis for relative strength trading 55 replies

Relative Strength Basket Trading System 3,479 replies

Dyekid217 Trading relative strength and CICR 0 replies

Attachments: Kang Gun Relative Strength trading - the EA's
Exit Attachments

Kang Gun Relative Strength trading - the EA's

  • #101
  • Feb 22, 2009 3:35am Feb 22, 2009 3:35am
I'm testing three different DC settings:
- FxOpen V5 Daily - Weekly result +26% (last week +5%)
- FXDD V7a Daily - Weekly result +4% (last week +15%)
- MIG V7a 4Hr - Weekly result -5%

Unfortunately I can only trade 3 platforms on the VPS. So what to test next week?

Variables:
- FXOpen indi value is more accurate (all the pairs), has no close on recross and also number of trades is a lot higher (prob due to the absence of the stoch filter)
- FXDD has the close on recross and approx half the trades of FxOpen
- Not really impressed with the 4hr approach - not enough reward for the risk - for the time being may discard

So, next week tests:
- FXOpen V1.3 Daily - With close on recross - same settings as FXDD but check the difference of the indi
- FXDD V1.3 Daily
- MIG V1.3 Daily - with stoch turned off

Cheers
  • #102
  • Feb 22, 2009 3:56am Feb 22, 2009 3:56am
Nice testing maxdd. Thanks for posting the results.

The logic behind DC order placement remains unchanged from the earliest versions of the EA, when it was a part of the monster robot that traded all three methods. Changes have mostly been adding more trading filters, so you can expect to see fewer trades but with lower draw down. That is the theory, at least.

FYI, in case you are not using it already, AccountEquityAnalyser is an indi posted by Toyoproxo in KG's thread. Amongst other things, it shows the maximum draw down the account has taken - useful if you are not able to monitor the accounts a lot. It is included in the zip file.

  • #103
  • Feb 22, 2009 4:12am Feb 22, 2009 4:12am
Thanks Steve for the info on the AccountEquityAnalyser - sounds useful.

Here's hoping for another good week!

Cheers
  • #104
  • Feb 22, 2009 4:16am Feb 22, 2009 4:16am
I just looked again at something I noticed when I was coding the Sahars method yesterday.

Once a cross has been validated, the final hurdle for the trade is the Stochastic filter. I see in the user guide that my original intention was merely to delay the trade if Stoch is in the overbought/sold areas. In fact, when the filter is against the trade, it is cancelled altogether and the cycle starts again.

A basic change to this would take seconds to code. In the case of a potential buy trade where Stoch indicates overbought, merely telling the EA to wait until Stoch levels drop will result in entering a trade too early.

Using the TradeKgMethod, the EA will take the buy just as the price is retracing. Maybe we need the EA to wait until Stoch has dropped, say, to Overbought (default 80) minus X, perhaps Overbought - 30? This approach risks missing the trade altogether if Stoch never falls that far. What do you guys think?

There are implications for the TradeSaharsMethod. Here is what I am thinking, looking again at a potential buy:

  1. Everything is lined up for sending a pending buy order, but Stoch indicates the market is overbought.
  2. Instead of sending a pending order that might get filled at an impossibly high Stoch setting, the EA waits for Stoch to drop back below the overbought level.
  3. Every 10 points (5?) that Stoch drops, the EA sends a pending order SaharsPendingDistancePips above the current market price, and deletes the previous order.
  4. Eventually, either a pending trade will fill, or there will be a recross which cancels everything so the process can start again.

Any thoughts about this, folks?

  • #105
  • Feb 22, 2009 4:38am Feb 22, 2009 4:38am
Go 1 time frame down and when that time frame turns back up, that will be signal to get in.
  • #106
  • Feb 22, 2009 5:47am Feb 22, 2009 5:47am
Quoting tfrench
Go 1 time frame down and when that time frame turns back up, that will be signal to get in.
And one to ponder. Thianks. Sometimes the simplest ideas are the best.

Talking of pondering, I have just had a bath. The significance of this seemingly pointless info is this: I ponder in the bath. For those of you whose first language is not English, 'pondering' is not some sort of kinky, perverted activity; it means 'thinking deeply'.

As a result of my ponder, V1.5 is on page 1. Read the brief details there.

  • #107
  • Feb 22, 2009 8:47am Feb 22, 2009 8:47am
Steve.

You are indeed a piano-teaching, pondering, coding machine.

Seriously, you probably get tired of reading it, but many thanks for all your tireless efforts and for your generosity in sharing.

saintmo
  • #108
  • Feb 22, 2009 9:25am Feb 22, 2009 9:25am
Interesting Thread, lots of good information... I'm am testing currently, but will post my results and things go on! Best wishes...
  • #109
  • Feb 22, 2009 4:17pm Feb 22, 2009 4:17pm
Cheers guys.

I forgot to say: I have coded 1.4 and 1.5 over the weekend and so have not been able to test the code. Please watch out for bugs that need squishing and post about them here.

  • #110
  • Feb 22, 2009 4:34pm Feb 22, 2009 4:34pm
Steve, thanks for all that you do.....

I have skipped 1.4 and now have 1.3 and 1.5 up on demos....

Is there a reason I should also try 1.4?

Wayne

Quoting SteveHopwood
Cheers guys.

I forgot to say: I have coded 1.4 and 1.5 over the weekend and so have not been able to test the code. Please watch out for bugs that need squishing and post about them here.

  • #111
  • Feb 22, 2009 5:14pm Feb 22, 2009 5:14pm
Quoting webicknell
Steve, thanks for all that you do.....

I have skipped 1.4 and now have 1.3 and 1.5 up on demos....

Is there a reason I should also try 1.4?

Wayne
Nope. !.5 extends but does not change 1.4, so demo 1.5.

  • #112
  • Feb 22, 2009 5:21pm Feb 22, 2009 5:21pm
Hello Steve and all:
I just loaded version 1.5 with daily chart and defaults..
I am getting the following:

18:02:37 KG Relative Strength Daily Cross Trading EA V1.5 GBPJPYm,Daily: Alert: EURUSDm DC Buy order send failed with error(130): invalid stops

Thanks
Vince
  • #113
  • Feb 22, 2009 5:39pm Feb 22, 2009 5:39pm
Hey guys. Thanks for posting.

webicknell has pm'd me with the same error. I have observed it as well on my demo.

I take it we are all running 1.4/1.5 with the default TradeSaharsMethod set to 'true'?

There is a bug in the code. Typically, code looks ok when I write it, but errors creep in that I do not spot, because I cannot test it before release over the weekend.

Right now, I am a few glasses into a nice bottle of red wine (OK, I confess, and followed up by a couple of whisky chasers) so bug-hunting is not going to be successful. I will sort this out tomorrow and release 1.6 with the bugs on the run at least, even if not entirely squished.

Watch this space, and please post if you are not running 1.4/1.5 with TradeSaharsMethodset to 'true', and still seeing this error message, as this would represent yet another bug.

Hic.

Cup.

  • #114
  • Feb 22, 2009 5:51pm Feb 22, 2009 5:51pm
Hello Steve and enjoy the wine,also I forgot to also thank Mr. Sahars in last post.
Vince
  • #115
  • Feb 22, 2009 6:04pm Feb 22, 2009 6:04pm
Quoting letizia
Hello Steve and enjoy the wine,also I forgot to also thank Mr. Sahars in last post.
Vince
Cheers. I am.

As soon as I looked at the code, I saw immediately what I had done wrong. Coders will understand my embarrassment when I confess to mixing up OP_BUYSTOP with OP_BUYLIMIT

Oops.

Putting this right still hasn't solved the problem, so the code is still not working. Ignore 1.4 and 1.5 for now, folks. I will hang onto 1.6 until I know the code works.

Sorry about this guys. The modifications to 1.3 seemed so simple when I made them. Clearly not. Hey ho.

  • #116
  • Feb 22, 2009 6:05pm Feb 22, 2009 6:05pm
Quoting maxdd
I'm testing three different DC settings:
- FxOpen V5 Daily - Weekly result +26% (last week +5%)
- FXDD V7a Daily - Weekly result +4% (last week +15%)
- MIG V7a 4Hr - Weekly result -5%

Unfortunately I can only trade 3 platforms on the VPS. So what to test next week?

Variables:
- FXOpen indi value is more accurate (all the pairs), has no close on recross and also number of trades is a lot higher (prob due to the absence of the stoch filter)
- FXDD has the close on recross and approx half the trades of FxOpen
- Not really impressed with the 4hr...
Thanks for the test results maxdd..

KG
  • #117
  • Feb 22, 2009 6:20pm Feb 22, 2009 6:20pm
There are times when I wonder how it is possible to be as dim as me and still continue as an extant life-form.

Even with a few drinkies inside me, I managed to spot the most glaring of errors in the code. V1.6 squishes the bugs and is now cheerfully placing the pending trades.

Not sure the defaults are the best ones, mind, but that is a different story. Contributions welcome here as usual about these.

  • #118
  • Feb 22, 2009 8:02pm Feb 22, 2009 8:02pm
Quoting SteveHopwood
V 1.4 is attached to the first post in this thread. There are some new inputs to handle a second DC trading method. From the user guide:...
Steve, Thanks for your time and effort to code this EA. I'll test it immediately.. Cheers Sahars
  • #119
  • Feb 22, 2009 8:39pm Feb 22, 2009 8:39pm
Quoting SteveHopwood
There are times when I wonder how it is possible to be as dim as me and still continue as an extant life-form.

Even with a few drinkies inside me, I managed to spot the most glaring of errors in the code. V1.6 squishes the bugs and is now cheerfully placing the pending trades.

Not sure the defaults are the best ones, mind, but that is a different story. Contributions welcome here as usual about these.


you funny, Steve-amigo. And goood work, with thanks.

Ron
  • #120
  • Feb 22, 2009 8:53pm Feb 22, 2009 8:53pm
Hi Steve,

Actually I just came back for long traveling. I thought V1.4 is the latest, but you are faster than my thought and now V1.6 is already released.

Now I'm testing V1.6... For the sake of testing, I would testing it in 2 platforms, one with Stoch enable and one with Stoch function disable.

Based on discussion with KG, somehow the EA was produced better result without the Stoch filter. Anyway, I will test it this week at default value (except on Stoch function) and post it the result daily for the discussion purpose. I attached here the set for my test value.

The idea of following the Stoch movement by adjusting the StopBuy value was a good idea.

Hopefully, by US session I could post the result for us to share and to improve it.

Again...Thanks Steve for your excellent coding...!!

Cheers

Sahars.
Attached File(s)
File Type: zip kg rs dc v1.6_stoch enabled.zip   1 KB | 594 downloads
File Type: zip kg rs dc v1.6_stoch disable.zip   1 KB | 528 downloads
0 traders viewing now