stevehopwoodforex.com
https://www.stevehopwoodforex.com/phpBB3/
Print view

Gday: a D1 Stochastic OB/OS trading robot
https://www.stevehopwoodforex.com/phpBB3/viewtopic.php?t=168		 Page 18 of 71
Author:  garyfritz [ Wed Dec 21, 2011 4:14 pm ]
Post subject:  Re: Gday: a D1 Stochastic OB/OS trading robot
Yes, if my tests are right, the average trade size on this is actually fairly reasonable -- 10-30 pip range. So we might be able to run it on shorter timeframes and add to the trade list.
Author:  RockyJ [ Wed Dec 21, 2011 4:47 pm ]
Post subject:  Re: Gday: a D1 Stochastic OB/OS trading robot
p.s. great work on fixing the jumping stop issue... ... so much brain power around here.. 8-)
It's impressive to see.
Author:  1of3 [ Wed Dec 21, 2011 5:19 pm ]
Post subject:  Re: Gday: a D1 Stochastic OB/OS trading robot
Steve, in the JumpingStopLoss() function you have:

Code: Select all

 [Line1] if (NewStop - OrderStopLoss() >= Point) modify = true;//George again. What a guy
[Line2] modify = true;

for the BUY and

Code: Select all

 [Line1] if (OrderStopLoss() - NewStop >= Point || OrderStopLoss() == 0) modify = true;//George again. What a guy   
[Line2]  modify = true;
for the SELL.

Does the second line of code not make the first line of code moot or have I misunderstood what it's doing?.
Author:  alex_forex [ Wed Dec 21, 2011 5:36 pm ]
Post subject:  Re: Gday: a D1 Stochastic OB/OS trading robot
My results for today (not all the good pairs were active tonight so let's see on the next start)

12.21.2011 14:21 USDCAD Sell 0.03 1.02919 0.97839 1.02939 1.02954 -1.5 -0.44 14h 21m

12.21.2011 13:23 AUDUSD Buy 0.03 1.01194 1.05784 1.00674 1.01194 52.0 15.60 13h 23m

Regards
Alex
Author:  gaheitman [ Wed Dec 21, 2011 5:49 pm ]
Post subject:  Re: Gday: a D1 Stochastic OB/OS trading robot
1of3 wrote:Steve, in the JumpingStopLoss() function you have:

Code: Select all

 [Line1] if (NewStop - OrderStopLoss() >= Point) modify = true;//George again. What a guy
[Line2] modify = true;

for the BUY and

Code: Select all

 [Line1] if (OrderStopLoss() - NewStop >= Point || OrderStopLoss() == 0) modify = true;//George again. What a guy   
[Line2]  modify = true;
for the SELL.

Does the second line of code not make the first line of code moot or have I misunderstood what it's doing?.
Yep, the second line shouldn't be there. I suspect the "modify = true" was visually lost on the first line so the second line is supposed to be the action to take if the "if" is true.

Good catch.
Author:  Gamma_gallus [ Wed Dec 21, 2011 5:57 pm ]
Post subject:  Re: Gday: a D1 Stochastic OB/OS trading robot
Hum I wonder how this puppy would behave if we switch on Hanover for W1 (maybe 1 slope confirmation) :?:

I'll see that in 2012 - well if I do remember ... - Off from trading for 2 weeks now.

I wish you all a Merry Christmas, enjoy the parties :D

Cheers
Gamma
Author:  SteveHopwood [ Wed Dec 21, 2011 6:18 pm ]
Post subject:  Re: Gday: a D1 Stochastic OB/OS trading robot
1of3 wrote:Steve, in the JumpingStopLoss() function you have:

Code: Select all

 [Line1] if (NewStop - OrderStopLoss() >= Point) modify = true;//George again. What a guy
[Line2] modify = true;

for the BUY and

Code: Select all

 [Line1] if (OrderStopLoss() - NewStop >= Point || OrderStopLoss() == 0) modify = true;//George again. What a guy   
[Line2]  modify = true;
for the SELL.

Does the second line of code not make the first line of code moot or have I misunderstood what it's doing?.
Nice one. I will fix it when I add the break even profit to JS later on.
Author:  SteveHopwood [ Wed Dec 21, 2011 7:27 pm ]
Post subject:  Re: Gday: a D1 Stochastic OB/OS trading robot
Latest update in post 1.

TIG ran his slide rule over the trailing stop stuff and made the same changes he made to JS.

I have reintroduced AddBEP to the JS function - I am not sure what I was thinking when I removed it from the shell. Hey ho.

So, BE and JS are now separate. BE is disabled by default; JS is enabled. Strictly speaking, BE is no longer needed but I leave it in for those who want a move to BE but then no further trailing/jumping of the stop.

I intended asking people if we might remove a few of the trend-spotting routines to shorten the inputs list, but I see people already starting to think about how to use the features. I shall leave them all in place.

I would be surprised to find that there is anything else in existing code to stop Gday working as intended - George is a very clever programmer. So, time to think about more flexible, market based sl/tp.

I think we have a few alternatives to play with:
  • 1) Standard manually chosen stop.
    2) My own volatility calculator. This works by working out the average daily pips movement over a set period of days. I prefer this to ATR only because I do not properly understand ATR.
    3) ATR
    4) 'Chandelier' stop as measured/indicated by PSAR.
Any thoughts, gals and galesses?

:D
Author:  SteveHopwood [ Wed Dec 21, 2011 7:33 pm ]
Post subject:  Re: Gday: a D1 Stochastic OB/OS trading robot
For anyone briefly confused, there was a typo my previous post. 'So, BE and JS are not separate. ' should have been, and is now, 'So, BE and JS are now separate. '

:D
Author:  alex_forex [ Wed Dec 21, 2011 7:34 pm ]
Post subject:  Re: Gday: a D1 Stochastic OB/OS trading robot
Thanks for this update ;)

Voting for solution number 2 in TP/SL calculation 8-)

Regards
Alex
SteveHopwood wrote:Latest update in post 1.

TIG ran his slide rule over the trailing stop stuff and made the same changes he made to JS.

I have reintroduced AddBEP to the JS function - I am not sure what I was thinking when I removed it from the shell. Hey ho.

So, BE and JS are now separate. BE is disabled by default; JS is enabled. Strictly speaking, BE is no longer needed but I leave it in for those who want a move to BE but then no further trailing/jumping of the stop.

I intended asking people if we might remove a few of the trend-spotting routines to shorten the inputs list, but I see people already starting to think about how to use the features. I shall leave them all in place.

I would be surprised to find that there is anything else in existing code to stop Gday working as intended - George is a very clever programmer. So, time to think about more flexible, market based sl/tp.

I think we have a few alternatives to play with:
  • 1) Standard manually chosen stop.
    2) My own volatility calculator. This works by working out the average daily pips movement over a set period of days. I prefer this to ATR only because I do not properly understand ATR.
    3) ATR
    4) 'Chandelier' stop as measured/indicated by PSAR.
Any thoughts, gals and galesses?

:D
All times are UTC Page 18 of 71

Powered by phpBB® Forum Software © phpBB Limited