Eurusd/gbpusd - page 448

 
firscall:
I'm not sure about next week. I have looked at the FX chart only once today!!!! Have a great weekend and chat to you next week!!!

good morning m8 ! and all . the no. for the week - gu 5675 ,5615 eu 3624 ,3564 have a nice week

 
avi1:
good morning m8 ! and all . the no. for the week - gu 5675 ,5615 eu 3624 ,3564 have a nice week

week no . 5615 buy -5648 tp 33p

 
avi1:
week no . 5615 buy -5648 tp 33p

week no 5615 from 5648-5682 tp2 34p

 
avi1:
good morning m8 ! and all . the no. for the week - gu 5675 ,5615 eu 3624 ,3564 have a nice week

week no. 5675 buy -5716 tp 41p

 
avi1:
week no 5615 from 5648-5682 tp2 34p

Morning m8!

I have spent the entire weekend attempting to make a system fail!!!!!

Files:
tsd.bmp  1849 kb
tsd2.bmp  406 kb
 
firscall:
Morning m8! I have spent the entire weekend attempting to make a system fail!!!!!

on what its baised?

 
firscall:
I can't find a problem. Seriously, I have tried 100s of things to make it fail but I can't do it. The system seems to work. The result above are based on out of sample data and consists of 2647 trades. The number of trades generated in sample was 15000.

nice i saw that its gave good results, but i dont get the no. ...generated in samole was 15000?

 
avi1:
on what its baised?

I can't find a problem. Seriously, I have tried 100s of things to make it fail but I can't do it. The system seems to work. The result above are based on out of sample data and consists of 2647 trades.

The number of trades generated in sample was 15000.

Initially, I was reviewing the performance data in Excel (I wanted to be able to see every assumption within a column) but the result were so breath-taking that I thought I had made a mistake somewhere. So I loaded the data into Matlab; again the results were too good so I used Market System Analyzer (I have used this for a number of years and it is quite robust) to look for holes in the method etc.

So I'm sat here sratching my head attempting to work out how I can make this system fail as all attempts have failed so far.

Files:
tsd3.jpg  98 kb
 
avi1:
nice i saw that its gave good results, but i dont get the no. ...generated in samole was 15000?

The 'in sample data' is the data that I used to contruct the model. Therefore the parameters of the model we're adjusted based on the structure of the 'in sample' data.

Good results 'in sample' doesn't mean that you'll enjoy good results in real life.

The out of sample data is the data that hasn't been seen by the model. Assuming that mistakes haven't been made, the out of sample data should be a reflection of actual system performance (all things remaining equal).

 
avi1:
good morning m8 ! and all . the no. for the week - gu 5675 ,5615 eu 3624 ,3564 have a nice week

week no. 3624 short 3590 tp 34p

Reason: