scooby-doo BOJ JPY pairs trading robot | Page 131 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Trading correlation pairs by using the other pairs 843 replies

Export All pairs or selected pairs to .csv with script 3 replies

Pairs of Currency Pairs 4 replies

Trending Pairs / Ranging Pairs 0 replies

Robot trading 120 replies

Attachments: scooby-doo BOJ JPY pairs trading robot
Exit Attachments

scooby-doo BOJ JPY pairs trading robot

  • #2,601
  • Nov 18, 2009 5:08am Nov 18, 2009 5:08am
Quoting scooby-doo
The only thing left todo is to somehow detect when there is a run on the Yen for some reason. This happend last week in the early hours (GMT) whereby the Bank of Taiwan tried too devalue JPY. Ofcourse they failed long term but in the short term it would have probably caused everyone's SL to be triggered.

This has nothing todo with economic news as such so it cant be easily spotted. But, i am hoping to discuss this with traders in Tokyo when I get there.

In case anyone is wondering.... My trading floor could have survived huge DD (to normal folk)...
I assume this sort of CB intervention does not happen very often - correct me if I am wrong.

Maybe we just have to accept that occasionally we will hit S/L?
  • #2,602
  • Nov 18, 2009 5:11am Nov 18, 2009 5:11am
Quoting SteveHopwood
I think I have found the source of the rogue trades this morning. The most likely possibility is that I forgot to rename the Overbought and Oversold globals in the 24H version.

What probably happened is this:
[list][*]Running both robots on the same demo.[*]24H version found the correct conditions to set an Oversold global during the 'official' Buy period..[*]'Proper' robot came across the global, found conditions just right for trading, yelled, "Yippee, some trades." and sent the Sell basket.[*]A couple of minutes later,...
yep - both robots on demo
  • #2,603
  • Nov 18, 2009 5:12am Nov 18, 2009 5:12am
Quoting maxou888
I renamed the globalvariables xxxxGv with xxxxxGV2 in the 24h version (v1.3). My surprise is when I check on the Mt4 platform tools>>GlobalVariables, I do not find any for this 2 Eas GV.
Very strange because the 2 ea working well : the 24h send 1 sell basket this morning, another sell basket yesterday evening, normal version didn't trade so far (I started this 2 bots yesterday afternoon)
The globals are only set when needed and are deleted when not. They come and go.

  • #2,604
  • Nov 18, 2009 5:18am Nov 18, 2009 5:18am
Quoting puntalara
yep - both robots on demo
Thanks. Looks like the globals thingy was the fault.

  • #2,605
  • Nov 18, 2009 5:18am Nov 18, 2009 5:18am
Quoting macman
I assume this sort of CB intervention does not happen very often - correct me if I am wrong.

Maybe we just have to accept that occasionally we will hit S/L?
Its a good question but one that I cant answer. It depends on different CB's policy at the time. Although there are no rules in the currency market some of these CB's do act against a so called gentleman's agreement which is not to try to upset other economies. But tell that to Mr W. Buffet who some years ago initiated a run on GBP, making $700 Million for himself and causing havoc for BOE.

That's another reason why I said about account size being >= 2500 given our base lot divisor. Then you would more than likely have survived the short term spike etc. Although you probably would have had a heart attack if you were watching your trades.
  • #2,606
  • Nov 18, 2009 5:21am Nov 18, 2009 5:21am
Quoting SteveHopwood
Updated the news filter code in both robots to include UJ.

Another logic fuzz guys. Help, coders. The robot needs to check, say, for GBP news. Can I simply send a call to FFCal Scoobs using GBPJPY as the parameter, or does it have to be specifically "GBP"? Or doesn't it matter?

Helllllllllllllllllllllllllppppppppppppppppp
The FFCal checks for news on the 2 currencies of the pair you use.
  • #2,607
  • Nov 18, 2009 5:25am Nov 18, 2009 5:25am
Quoting scooby-doo
In case anyone is wondering.... My trading floor could have survived huge DD (to normal folk) as it would have been insignificant on the scale of things and the system would have hedged the carry trade until the Yen restored its value. This would be difficult for retail traders todo as you would have to pay and overcome a double spread.

Scoobs.
Hedging can be built into the robot if you can suggest a strategy. I coded hedging routines into a lot of bots, but the technique was probably a bit unsophisticated. Users of my latest AshFX bot will be familiar with this.

  • #2,608
  • Nov 18, 2009 5:26am Nov 18, 2009 5:26am
Quoting challenger78
The FFCal checks for news on the 2 currencies of the pair you use.
Cheers.

  • #2,609
  • Edited 5:42am Nov 18, 2009 5:32am | Edited 5:42am
Quoting SteveHopwood
Hedging can be built into the robot if you can suggest a strategy. I coded hedging routines into a lot of bots, but the technique was probably a bit unsophisticated. Users of my latest AshFX bot will be familiar with this.

Because of cross rates etc then you would hedge EURJPY and USDJPY.

So, basically if we SELL XXXJPY and there is a run on JPY then you would BUY EURJPY and USDJPY to help negate your DD. There is no point sending all BUY XXXJPY pairs as when things turn around via BOJ intervention you would end up in no mans land.

You hedge to reduce DD in the short term until things turn around.

What you could do:

If there is no direct economic news affecting the basket and you hit 5% DD (probably a run on the Yen or major disaster affecting Asia Pacic region) then hedge EURJPY and USDJPY.

You would use the same lots for each pair that you sent previously.
  • #2,610
  • Nov 18, 2009 5:47am Nov 18, 2009 5:47am
Quoting SteveHopwood
Lot for lot?

You would use the same lots for each pair that you sent previously.
  • #2,611
  • Nov 18, 2009 5:48am Nov 18, 2009 5:48am
Quoting scooby-doo
You would use the same lots for each pair that you sent previously.
Sorry. Just realised I had not read all of the post. Cheers.

  • #2,612
  • Nov 18, 2009 5:51am Nov 18, 2009 5:51am
Quoting scooby-doo
Because of cross rates etc then you would hedge EURJPY and USDJPY.

So, basically if we SELL XXXJPY and there is a run on JPY then you would BUY EURJPY and USDJPY to help negate your DD. There is no point sending all BUY XXXJPY pairs as when things turn around via BOJ intervention you would end up in no mans land.

You hedge to reduce DD in the short term until things turn around.

What you could do:

If there is no direct economic news affecting the basket and you hit 5% DD (probably a run on the Yen or major disaster affecting Asia...
This gives me something else to think about.

When do we close the hedge?

  • #2,613
  • Edited 6:04am Nov 18, 2009 5:52am | Edited 6:04am
Quoting SteveHopwood
Sorry. Just realised I had not read all of the post. Cheers.

...before anyone asks.... what happens if things dont turn around and the basket cant hit TP ?

Then it would require you to manually intervene in someway.

I know people want to code in every possible scenario etc but that just doesnt happen in the real world. Again, if it did, then I would be out of a job.

Example:

You hit 5% dd = then hedge. Things turn around and you now have only 1% dd. You manually decide if you want to see if it will hit your TP or any TP or close out the small lose.

Its a decision for the user/trader and not a Robot running a predefined set of what/if rules.

And dont forget that this is worse case scenario!!!

PS. if things dont turn around then you keep hedging and close out when your profit/loss is as close to breakeven as possible.
  • #2,614
  • Nov 18, 2009 6:06am Nov 18, 2009 6:06am
Quoting scooby-doo
...before anyone asks.... what happens if things dont turn around and the basket cant hit TP ?

Then it would require you to manually intervene in someway.

I know people want to code in every possible scenario etc but that just doesnt happen in the real world. Again, if it did, then I would be out of a job.

Example:

You hit 5% dd = then hedge. Things turn around and you now have only 1% dd. You manually decide if you want to see if it will hit your TP or any TP or close out the small lose.

Its a decision for the user/trader and not...
Yes. I am not sure I am going to do anything about it. I trust you when you say how accurate this system is, so I don't think it is necessary. I suppose I am thinking out loud, really.

  • #2,615
  • Nov 18, 2009 6:11am Nov 18, 2009 6:11am
Quoting SteveHopwood
Yes. I am not sure I am going to do anything about it. I trust you when you say how accurate this system is, so I don't think it is necessary. I suppose I am thinking out loud, really.

Steve,

Its why I said balance needs to be >= 2500, otherwise if there is a run on the Yen and you dont build in hedging etc then you will hit SL.

So, maybe you should build it in incase this scenario occurs and people dont have enough equity to ride it out until probable turn around.
  • #2,616
  • Nov 18, 2009 6:13am Nov 18, 2009 6:13am
Hi guys, still following this thread with great interest.
Scooby-Doo, could I pick your brain regarding the following please? You have mentioned a while back you look at the distance the price is away from the 200SMA to gauge what is going on. What time frame do you use for this please?

Cheers,

Vince
  • #2,617
  • Nov 18, 2009 6:19am Nov 18, 2009 6:19am
>> Scooby-Doo
I think about hedging strategy that I do not understand because I probably not using the right technique.
Each time i use hedge when a DD occurs, if the hedge size is the same size that the initial trade, I always keep a loss that is the gap between the initial trade price and the hedge price, no matter in witch direction the price move after the hedge have been initiated.
Choosing to not close the both sides at the same moment is a way, but also a bet on the next future move, could be better, could be worse.
Hedging with another pair, same dilema, can brings the things better, but also worsening.
Is there a clever style for hedging ?

Thanks
  • #2,618
  • Nov 18, 2009 6:26am Nov 18, 2009 6:26am
Quoting VinceTheBeast
Hi guys, still following this thread with great interest.
Scooby-Doo, could I pick your brain regarding the following please? You have mentioned a while back you look at the distance the price is away from the 200SMA to gauge what is going on. What time frame do you use for this please?

Cheers,

Vince
The next higher timeframe > x3. So if M15 look at 1H etc
  • #2,619
  • Nov 18, 2009 6:27am Nov 18, 2009 6:27am
Quoting scooby-doo
The next higher timeframe > x3. So if M15 look at 1H etc
Thanks
  • #2,620
  • Nov 18, 2009 6:30am Nov 18, 2009 6:30am
Steve, I also had the rogue trade with both robots running on the same demo account. Thanks for all your work.

Jack
12 traders viewing now