Thanks very much for sharing the info
JR,
I usually modifiy the settings by running a few backtests and sometimes the MT4 Optimisation, but that usually is pretty computer intensive.
Usually some educated guesses on what to change in the settings and a few backtests is all that's necessary to update the EA to give you some better results according to the current market conditions.
Marci
Presumably v200 is the best, and contains in itself the best settings? In post 1 you imply that it is still buggy.
Have you ever correlated backtest vs forward results for any Flyer version, for a period of at least a few weeks?
Flyer is interesting in that it is the first EA I've found which tries to react to the market. All others have been ambush-type i.e. expect a certain market behaviour like the London breakout and set trap accordingly.
Hope you don't mind all the Qs
(is there a FAQ, can't seem to see one)JRtrader
Darkness lies one inch ahead..
Yes, I tried optimising Flyer v1.05A on my Athlon XP 3200+ 1 GB. It took 10 hr 45 min to compute 124 of 1360 parameter sets, at which point I aborted (estimated total time 107 hr) and of the 124, all but 8 were $0 profit. The 8 were all losses, with drawdown ranging from $2257.60 thru $5947.93. Of course the full set might have given some winners(?)
That's OK if you're experienced, but otherwise... How far do you look back?
Presumably v200 is the best, and contains in itself the best settings? In post 1 you imply that it is still buggy.
Have you ever correlated backtest vs forward results for any Flyer version, for a period of at least a few weeks?
Flyer is interesting in that it is the first EA I've found which tries to react to the market. All others have been ambush-type i.e. expect a certain market behaviour like the London breakout and set trap accordingly.
Hope you don't mind all the Qs
JRtrader
Sorry for the delay in getting back to you as I'm not always on the forum daily these days. May get back to that soon, but other matters press at the moment.
V2.00 is my most recent version and the most reliable in my opinion. I think I posted a minor update in which I did not change the version number which fixed a problem with the EA trading during hours it wasn't supposed to when the LimitHours variable is set.
I have not updated the settings recently for the EA as I have not been using it as much in my personal trading. I know some people have been getting decent results with the EA and presumably they have updated it accordingly as I've had few posts regarding such.
As to updating the settings, you only need to run the backtests back to a point in time at which the EA is not being profitable. So if it stopped being regularly profitable on say Feb 1st, then backtest it maybe through Jan 1st and see what settings give you a profitable run. The optimisation routines on MT4 have to be restricted to only test a small range of parameters in which you think might give you good results (ie only test 20-30 combinations). Otherwise forget it unless you want to wait a week to get your results.
I have backtested it over periods in which it has been forward tested and I feel that the results pretty much are the same. However you have to be careful that you start the backtest at exactly the same time you started your forward testing (yes starting times make a difference in EA performance).
Hope all this helps you out. I may get around soon to posting some new updates and improvements to the EA as I enjoy helping people improve their trading. Learning how to trade using a moderately sucessful EA is one tool to
help you, but don't rely upon it. Take the time to sit down and find out why the EA works as well as it does (or doesn't) and it will help you make decisions as well.
Good Luck
Marci (GirlFlyer)
- #129
- Apr 30, 2007 6:57pm Apr 30, 2007 6:57pm
Yes, once I have it finished. I've started work on it already, but this one will be more difficult to program than Flyer! was so it may be a month or so before I have a preliminary version available to try out.
cheers
TraderJR
Darkness lies one inch ahead..
