Menu
Sister Sites

Options

Bookmark Thread

First Page First Unread Last Page Last Post

Print Thread

Similar Threads

Trading correlation pairs by using the other pairs 729 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

Quoting scooby-doo
I shall probably get banned soon as their was a thread where they were knocking british traders. I publicly told them what I thought.

However, I shall still read the thread and keep in touch with Steve via external email.

All the best

Scoob.
We would prefer you didn't get banned. Can't you quietly remove the offending posts? You are one of the most useful posters around, after all.

Quoting SteveHopwood
I have put the new versions with enhanced hedging techniques into post 1 so the brains around here can check the code. The relevant functions are:

 

  1. void CheckBasketTpSlTarget(): includes the code that takes the decision that hedging is required
  2. bool SendHedgeTrades() handles the actual hedging

You are dead right, scoobs. The code is a piece of piss but the logic would have taken me a while to get around, so you saved me some brain pain there. Cheers.

Actually, what takes most of the time is: copying the changes to the second...

5% could actually be too low as these XXJPY pair's are far ranging, looking at Friday afternoon one set of trades went from -7400 to + 2800 in an hour, although Friday's trading can be strange at the best of times.
Quoting scooby-doo
5% could actually be too low as these XXJPY pair's are far ranging, looking at Friday afternoon one set of trades went from -7400 to + 2800 in an hour, although Friday's trading can be strange at the best of times.
It seems low to me, but I am not going to argue. 'Tis but the adding of a single word to the code to make the variable a user input so people can experiment. What do you think?

Quoting SteveHopwood
It seems low to me, but I am not going to argue. 'Tis but the adding of a single word to the code to make the variable a user input so people can experiment. What do you think?

Yes fine.

However, you no what will happen. Some will set it to 25% and with a small account will get margin called. But thats life.

What I would do is, if Demo() then they can set it to what they like, if Live then it must kick in no higher than 7%.

Scoob.
Quoting scooby-doo
Yes fine.

However, you no what will happen. Some will set it to 25% and with a small account will get margin called. But thats life.

What I would do is, if Demo() then they can set it to what they like, if Live then it must kick in no higher than 7%.

Scoob.
Will do.

Quoting SteveHopwood
Will do.

I know you said it should be only run on demo but I think some are running it on live account. Which is a worry as its not yet finished.

But, hey ho.
Steve

I noticed like others did that there is a problem with the indicator when there is a new bar. It would be more accurate if someone could fix this although I must admit I dont have a clue how too.

Scoob.
Quoting scooby-doo
I know you said it should be only run on demo but I think some are running it on live account. Which is a worry as its not yet finished.

But, hey ho.
Most likely the more experienced people are doing this; they are able to remove the code that stops the 24H bot being run live.

I am also going to add TradeLong and TradeShort to the 24H, so people with different trend-spotting methods have more control.

Which reminds me, the bot also wants a Friday close and some dead-time hours imputs.

Or are dead-time inputs really necessary? Is there likely to be enough movement in the markets to trigger a basket but insufficient to then move it to tp? What do you think?

Quoting SteveHopwood
Most likely the more experienced people are doing this; they are able to remove the code that stops the 24H bot being run live.

I am also going to add TradeLong and TradeShort to the 24H, so people with different trend-spotting methods have more control.

Which reminds me, the bot also wants a Friday close and some dead-time hours imputs.

Or are dead-time inputs really necessary? Is there likely to be enough movement in the markets to trigger a basket but insufficient to then move it to tp? What do you think?

When London closes there is a big drop in volumes. For this reason I would stop trading at 4.00pm GMT and start again at 9.00pm GMT.

Personally, Friday close should be 12.00pm GMT but I am not going to get into an argument about all the reasons for that.

One of the biggest issues is around the indicator and when new bars happen.

Scoob.
Steve

I have come up with a very crude hack for the indicator problem whereby not all the new bars have formed at the same time, basically we wait 2 seconds before getting new values:

for(int i = 0; i < limit; i++)
{

if(Current.Bar != Bars)
{
Current.Bar = Bars;
Sleep(2000);
}

aud = 100 - iRSI("AUDJPY"+Suffix,0,2,PRICE_WEIGHTED,i);
cad = 100 - iRSI("CADJPY"+Suffix,0,2,PRICE_WEIGHTED,i);
chf = 100 - iRSI("CHFJPY"+Suffix,0,2,PRICE_WEIGHTED,i);
eur = 100 - iRSI("EURJPY"+Suffix,0,2,PRICE_WEIGHTED,i);
gbp = 100 - iRSI("GBPJPY"+Suffix,0,2,PRICE_WEIGHTED,i);
nzd = 100 - iRSI("NZDJPY"+Suffix,0,2,PRICE_WEIGHTED,i);
usd = 100 - iRSI("USDJPY"+Suffix,0,2,PRICE_WEIGHTED,i);
Buffer[i] = (aud + cad + chf + eur + gbp + nzd + usd) / 7;

}
Quoting scooby-doo
Steve

I have come up with a very crude hack for the indicator problem whereby not all the new bars have formed at the same time, basically we wait 2 seconds before getting new values:...
Great. I will pop that into the indi and people can re-download it in time for Sunday night. It isn't a problem for the robot as the calculation code is encapsulated within, so a bit crude doesn't really matter.

Quoting SteveHopwood
Great. I will pop that into the indi and people can re-download it in time for Sunday night. It isn't a problem for the robot as the calculation code is encapsulated within, so a bit crude doesn't really matter.

Its still a problem for the encapsulated code aswell, it needs to wait for all the new bars to have formed before caluclation RSI's etc.

However, I dont really know how long it takes for all the bars to form properly, maybe it is a matter of milliseconds and 2 seconds is too long to wait.

Maybe someone else can comment on this ? i.e. should we wait 1, 2, 3 seconds ?

Steve, you need to put Current.Bar = Bars; in the init()

Scoobs

PS. This is one thing that makes the indicator spike sometimes.
Hi guys
I had some spare time and made a dashboard for those trading this manually... maybe it helps

Showing the price above below 200SMA in 3 timeframes and the 15M and H1 Strength reading with a small buy/sell/wait indi.

Here's how it looks like.
Indi attached. Can't test it as it needs ticks...

Update 22.11.2009:
It's standalone now, no other indicator needed
It does not take CAD and NZD in the calculations. If you need or want it... tell me!
Showing recommended Base Lot Size.
Attached Image (click to enlarge)
Click to Enlarge Name: dxt.gif Size: 21 KB
Attached File
File Type: ex4 DXT Scooby.ex4   14 KB | 233 downloads
"The only successful substitute for brains is silence."
Quoting DXTrade
Hi guys
I had some spare time and made a dashboard for those trading this manually... maybe it helps

You still need the RS_GROUP_JPY Strength Index.ex4 in your indicator folder (not on the chart) but later this week I will try to include it.

Showing the price above below 200SMA in 3 timeframes and the 15M and H1 Strength reading with a small buy/sell/wait indi.

Here's how it looks like.
Indi attached. Can't test it as it needs ticks...
Cheers Uli; that's brilliant.

There is a limit of 7 uploads to a post, so it can't go in post 1 for now. When a couple of hours of ticks have satisfied you that it works, I will put a link to it in post1.

Quoting scooby-doo
Its still a problem for the encapsulated code aswell, it needs to wait for all the new bars to have formed before caluclation RSI's etc.

However, I dont really know how long it takes for all the bars to form properly, maybe it is a matter of milliseconds and 2 seconds is too long to wait.

Maybe someone else can comment on this ? i.e. should we wait 1, 2, 3 seconds ?

Steve, you need to put Current.Bar = Bars; in the init()

Scoobs

PS. This is one thing that makes the indicator spike sometimes.
The robot waits for 30 seconds at the start of each new candle. It would be good to get a more accurate setting for this.

Quoting SteveHopwood
Cheers Uli; that's brilliant.

There is a limit of 7 uploads to a post, so it can't go in post 1 for now. When a couple of hours of ticks have satisfied you that it works, I will put a link to it in post1.

Thanks Steve
but wait a bit. Need some time to check it and want to integrate the RSI in the indi as well. And maybe some cosmetics once it's running.
I might be as you and make some 1,435 versions before it's perfect
"The only successful substitute for brains is silence."
I just realised the sleep for 30 seconds function is not in the 'proper' version. I will add it.

Quoting SteveHopwood
I just realised the sleep for 30 seconds function is not in the 24H version. I will add it.

30 seconds!!!!!!!!!!!!!!!!!!!!!

Jesus Steve, i dont think it needs to wait that long????????????????????
Quoting scooby-doo
30 seconds!!!!!!!!!!!!!!!!!!!!!

Jesus Steve, i dont think it needs to wait that long????????????????????
5 seconds? There are times when we can wait for what feels like a mini ice age for a tick.

Quoting SteveHopwood
5 seconds? There are times when we can wait for what feels like a mini ice age for a tick.

3 seconds should be ample on a live account for all 7 bars to form unless the broker is a crook and delaying prices!!!!!!!!!!!!!!!!!
0 traders viewing now