jsidhu PM me. I have some experience with Neural nets, but think they are hard to use (except very simple perceptrons) in MT4 without a DLL. I have modified the public version of Thunder (v1e) so that it works well, and will share the source, but would like to improve for other pairs. I cannot PM for some reason... maybe the moderator dislikes me?

Yes there are doubts. Pls. read the thread especially the last 2 or 3 weeks. Your account is new perhaps you didn't have enough time to read it all. What do you mean with "hunting a possible non-freeware" ?? I'm hunting nothing and the EAs I'm using wiht my live-account are commercial ones and correctly registered.

I think, rdb is NanjanaFX and he is watching us.

Yes i have use them, and a lot of my friends too, you can contact me via YM, my YM is : machsus.ma@yahoo.com. Thanks

Madcow,


I'm not sure how to PM you I don't see an option under your profile either! Just email me at sidhujag@hotmail.com.

Regarding NN's, why not use DLL's? I have some examples doing this with MT4. You can write a script to optimize your inputs aswell by running multiple MT4 shells. Anyways the design is most important... what do you want to model, trends/news or what? with trend and news it may be too many inputs... but thats the reality of the situation, it IS complex. The best thing would be to keep optimizing these inputs, and even add/remove inputs by testing them under mt4 optimization to see what works best every day. That way the EA is updated to model the pattern of the currency as time goes.

Jag

reallly want to know this EA to be developed further, my demo make profit everyday even when i set it to trade only 3 hours (i avoid the asian open market to be safer :P) SO i trade when ALL market are closed

Hey EZY, if its password protected, how did you manage to do the back test? I am also after version 2e.

good trade with yourlucky v2b at alpari UK demo
Eurchf is the profit gainer, not eurgbp though.
^^

Hello to everybody here.

I did again a lot of backtesting with a lot of settings, and the result is:
Thunder, Lucky and even Fap is an useless EA if you choose a broker with 5 or 4 pip spread. It will kill every equity.

And there is no serios Broker out there who is offering 3 pip spread. forget about demotesting, and maybe live results with a 5 pip spread broker.

the only way is to find a non bucket shop broker who offering 3 pip spread on eur gbp

any ideas here ???

What helps us any discussion and talking, if we are not able to go live ? thats the brokers game and the gains are on the brokers side... sad but true

P.s: i did some live trading too, and it was a bit successful but its like a ticking bomb i guess.

results (same settings same broker same time period only difference live 5 pip spread, demo 3 pip spread)

1st demo looks very good
Bars in test 14645 Ticks modelled 908367 Modelling quality n/a
Mismatched charts errors 167
Initial deposit 10000.00
Total net profit 38605.39 Gross profit 78086.00 Gross loss -39480.61
Profit factor 1.98 Expected payoff 84.11
Absolute drawdown 338.59 Maximal drawdown 10600.24 (25.70%) Relative drawdown 25.70% (10600.24)
Total trades 459 Short positions (won %) 195 (84.10%) Long positions (won %) 264 (81.06%)
Profit trades (% of total) 378 (82.35%) Loss trades (% of total) 81 (17.65%)
Largest profit trade 718.35 loss trade -1954.69
Average profit trade 206.58 loss trade -487.41
Maximum consecutive wins (profit in money) 28 (7559.29) consecutive losses (loss in money) 4 (-3532.14)
Maximal consecutive profit (count of wins) 7559.29 (28) consecutive loss (count of losses) -5454.52 (3)
Average consecutive wins 7 consecutive losses 2


now live result not nearly that safe and good.

Bars in test 14361 Ticks modelled 891451 Modelling quality 90.00%
Mismatched charts errors 0
Initial deposit 10000.00
Total net profit 810.88 Gross profit 23863.01 Gross loss -23052.14
Profit factor 1.04 Expected payoff 1.95
Absolute drawdown 2633.47 Maximal drawdown 4130.28 (35.93%) Relative drawdown 35.93% (4130.28)
Total trades 416 Short positions (won %) 182 (79.12%) Long positions (won %) 234 (70.94%)
Profit trades (% of total) 310 (74.52%) Loss trades (% of total) 106 (25.48%)
Largest profit trade 162.57 loss trade -521.55
Average profit trade 76.98 loss trade -217.47
Maximum consecutive wins (profit in money) 23 (2040.91) consecutive losses (loss in money) 4 (-1670.23)
Maximal consecutive profit (count of wins) 2040.91 (23) consecutive loss (count of losses) -1670.23 (4)
Average consecutive wins 5 consecutive losses 2

Reduce work hours to get optimal performance, reduce about 3 hours in all broker, so only work 4 hours. If the works hours is 22-5 make it 22-2. And don't forget to play with broker which have 3 spread maximum in EUR/GBP. thanks