EAs under construction II

The tests are for Jan10 to July10
and July09 to June10. They were run on AAAFX and the tester doesn't seem to want to go past June. I've downloaded from history. It just doesn't want to go any further.
It's just the big stop loss and small gains that causes the problem.

all it takes is one big stop loss to be hit and a year or more of gains is gone....
this is what is referred to ask risk/reward ratio. The best strategy is the one that the reward is greater than the risk...

The reverse settings we are working on will do that. Make that risk reward ratio practical.

There should be some idea of a strategy. Otherwise as Taiyakixz has pointed out to me many times, testing and adjusting are just curve fitting exercises.

We are all learning together.
If we don't make some effort how can our learning process take place? Large part of learning is learning from mistakes, errors, misconceptions, etc. If we knew everything already there would be no meaning to the word learning.

These test used your original settings from earlier post.
USDJPY 4hr AAAFX

A note about the reverse settings we are working on.
If results are consistent then that is desirable. If a setting produces consistent winnings that is good. But if a setting produces consistent losses that is also good. It is possible to trade opposite.
It is the consistency that is wanted.

A little more about the inverse settings.

This setting provides consistent losses on many different pairs.

NZDUSD here is a good example of the scalping results on the 5 min.
Spread is considered in calculating the profit and there is no rollover. The test was done on AAAFX spreads vary in live trading around 4 pips...
This is scalping setting and will provide 8 pips per trade. Stop loss is 12 and tgt1 and tgt2 are both 10 pips... 1 micro lot....
Talk about consistent........................................
AAAFX tester won't give good modeling quality. I run the same tests on fxdd to compare. Fxdd will give 90%.

SymbolNZDUSD (New Zealand Dollar vs US Dollar) Period5 Minutes (M5) 2010.01.03 23:00 - 2010.07.23 07:00 (2010.01.01 - 1970.01.01) ModelEvery tick (the most precise method based on all available least timeframes) ParametersiMagicNumber=8435701; GMTOffset=3; TradeStartHour=0; TradeEndHour=24; Lots=0.1; TakeProfit1=10; TrailingStop1=20; TakeProfit2=10; TrailingStop2=20; StopLoss=12; PendingPips=45; ExpireCandles=25; DelayOrder2=20; CheckSlope=false; DSS_name="Robby DSS Bressert Colored with alert v1.1"; DSS_EMA=8; DSS_stoch=13; Use_T3MA=false; T3MA_name="T3MA"; T3MA_Periods=0; T3MA_VolumeFactor=0.7; Channel_EMAPeriod=55; Channel_EMAShift=0; ShowComment=true; EXP_MA_1="-------- MA settings --------"; EXP_MA_2="-- Method 0:SMA, 1:EMA, 2:SMMA, 3:LWMA"; EXP_MA_3="-- Timeframe 0:current or minutes"; Use_MA1=false; MA1_Timeframe=0; MA1_Period=0; MA1_Method=1; MA1_Shift=0; Use_MA2=false; MA2_Timeframe=0; MA2_Period=0; MA2_Method=1; MA2_Shift=0; Use_MA3=false; MA3_Timeframe=0; MA3_Period=0; MA3_Method=1; MA3_Shift=0; Use_MA4=false; MA4_Timeframe=0; MA4_Period=0; MA4_Method=1; MA4_Shift=0; Use_MA5=false; MA5_Timeframe=0; MA5_Period=0; MA5_Method=0; MA5_Shift=0;
Bars in test24630Ticks modelled2074812Modelling qualityn/a Mismatched charts errors126




Initial deposit10000.00



Total net profit-466.00Gross profit50.00Gross loss-516.00 Profit factor0.10Expected payoff-9.71

Absolute drawdown466.00Maximal drawdown466.00 (4.66%)Relative drawdown4.66% (466.00)
Total trades48Short positions (won %)28 (14.29%)Long positions (won %)20 (5.00%)
Profit trades (% of total)5 (10.42%)Loss trades (% of total)43 (89.58%) Largestprofit trade10.00loss trade-12.00 Averageprofit trade10.00loss trade-12.00 Maximumconsecutive wins (profit in money)2 (20.00)consecutive losses (loss in money)13 (-156.00) Maximalconsecutive profit (count of wins)20.00 (2)consecutive loss (count of losses)-156.00 (13) Averageconsecutive wins1consecutive losses9

hi jiva34.

i sent you a pm. hope you can help me with some simple EA programming. TQ.

v7.0, GBPUSD H1 2009.08.03 - 2010.07.13
iMagicNumber=8435701; GMTOffset=2; TradeStartHour=0; TradeEndHour=24; Lots=0.1; TakeProfit1=30; TrailingStop1=80; TakeProfit2=1000; TrailingStop2=100; StopLoss=1000; PendingPips=0; ExpireCandles=10; DelayOrder2=20; CheckSlope=true; DSS_name="Robby DSS Bressert Colored with alert v1.1"; DSS_EMA=8; DSS_stoch=13; Use_T3MA=false; T3MA_name="T3MA"; T3MA_Periods=8; T3MA_VolumeFactor=0.7; Channel_EMAPeriod=38; Channel_EMAShift=0; ShowComment=false; EXP_MA_1="-------- MA settings --------"; EXP_MA_2="-- Method 0:SMA, 1:EMA, 2:SMMA, 3:LWMA"; EXP_MA_3="-- Timeframe 0:current or minutes"; Use_MA1=false; MA1_Timeframe=0; MA1_Period=5; MA1_Method=1; MA1_Shift=0; Use_MA2=false; MA2_Timeframe=0; MA2_Period=12; MA2_Method=1; MA2_Shift=0; Use_MA3=false; MA3_Timeframe=0; MA3_Period=20; MA3_Method=1; MA3_Shift=0; Use_MA4=false; MA4_Timeframe=0; MA4_Period=55; MA4_Method=1; MA4_Shift=0; Use_MA5=false; MA5_Timeframe=0; MA5_Period=197; MA5_Method=0; MA5_Shift=0;



But I will not trade with such settings....
weekend spread is large, so take care when you do backtest...

plugging away at these inverse settings:
nice consistency for EURUSD..
Can't seem to get my FXDD tester to run for more than a few trades. Apologize for the n/s but that's all I can get out of AAAFX. Will have to shop around some more for a platform that can give acceptable results. But at least we can see from the opening and closing of the trades given the data that it has been given, the ea is very consistent...
The time frame was set for about 1 year, the account was fully drained in 1 month......

SymbolEURUSD (Euro vs US Dollar) Period5 Minutes (M5) 2009.08.03 00:00 - 2010.05.07 20:55 (2009.08.03 - 2010.07.07) ModelEvery tick (the most precise method based on all available least timeframes) ParametersiMagicNumber=8435701; GMTOffset=0; TradeStartHour=1; TradeEndHour=24; Lots=1; TakeProfit1=10; TrailingStop1=20; TakeProfit2=10; TrailingStop2=20; StopLoss=10; PendingPips=0; ExpireCandles=25; DelayOrder2=0; CheckSlope=false; USE_DSS_name=false; DSS_name="Robby DSS Bressert Colored with alert v1.1"; DSS_EMA=8; DSS_stoch=13; Use_T3MA=false; T3MA_name="T3MA"; T3MA_Periods=0; T3MA_VolumeFactor=0.7; Channel_EMAPeriod=55; Channel_EMAShift=0; ShowComment=true;
Bars in test54823Ticks modelled8210640Modelling qualityn/a Mismatched charts errors330

more on these inverse settings
instead of buy a sell is placed.
instead of a sell a buy is placed.
no pending orders, market orders.
this is usdcad with a 2ema channel
20 stop loss, 10 take profit.
tester is set for 3 mos.
account drained in 1 month.

will wonders ever cease? 90% modeling on AAAFX...

the ea is not coded for this as of present, so in the test we are considering a loss as win and a win as loss.....

SymbolUSDCAD (US Dollar vs Canadian Dollar) Period5 Minutes (M5) 2010.04.01 00:00 - 2010.05.07 20:55 (2010.04.01 - 2010.07.07) ModelEvery tick (the most precise method based on all available least timeframes) ParametersiMagicNumber=8435701; GMTOffset=0; TradeStartHour=1; TradeEndHour=24; Lots=1; TakeProfit1=10; TrailingStop1=20; TakeProfit2=15; TrailingStop2=20; StopLoss=20; PendingPips=0; ExpireCandles=25; DelayOrder2=0; CheckSlope=false; USE_DSS_name=false; DSS_name="Robby DSS Bressert Colored with alert v1.1"; DSS_EMA=8; DSS_stoch=13; Use_T3MA=false; T3MA_name="T3MA"; T3MA_Periods=0; T3MA_VolumeFactor=0.7; Channel_EMAPeriod=2; Channel_EMAShift=0; ShowComment=true;
Bars in test8678Ticks modelled755004Modelling quality90.00% Mismatched charts errors0

GBPJPY with the same settings as USDCAD wiped an account out in a week!!!
Think inverse....
This makes the dreaded consolidation a dream come true....
Spread needs to be accounted for since profit will be minus the the spread.
FXDD is fixed spread so that makes it a little easier to calculate.

I want to hear the other member's opinion...
Anyone else has the idea of jiva's strategy?

He told me that the strategy of v7 was profitable. I believed it and made v7.
But the next day after I made it, he started to think of swapping buy and sell...

He set t/p=10 and s/l=20. If inverse, this means that s/l=10 and t/p=20.
With the broker I use, spread of gbpjpy is 5.5.
This means that 4.5pips moving reaches s/l, and that 25.5pips moving required to reach t/p.
Can we think that it's profitable...?

Of course making the ea and testing is the best way.
But already I modified the ea more than 30 times(included bug fix, so not all is caused by changing his idea, but...), I want to avoid to make no meaning ea if possible...

In reply to Taiyakixz frustration:
We are grateful for your hard work.
There are various applications for this strategy. We are exploring them.
At present a scalping application has been revealed. Which requires inverse settings. Buy signals are treated as sell signals and sell signals are treated as buy signals.
Since scalping means short trades, 15 pips or so the spreads become very significant for profits.
For different brokers there will be different spreads.
I am trading this in inverse at present and it is working very nicely.
For spreads of:
..........limit(loss) ....stop(profit)....channel.....
2 pips ...... 15.......... 15.................2 (calculates to 13 gain, 17 loss..there are more gains than losses)
3 pips .......15 ..........15.................2 (calculates to 12 gain, 18 loss..still profitable more gains than losses)
4 pips........15...........15.................2 (calculates to 11 gain, 19 loss..still profitable more gains than losses)
0 pending, dss set to false
more than 4 pips spread is not practical for this scalping application.
I am trading live FXDD: EURJPY, GBPUSD 4 pips; USDJPY, AUDUSD 3 pips; EURUSD 2 pips, and a special setting for USDMXN (slower channel and bigger stop loss)

I am not a programmer. But I have tried to modify Taiyakixz"s v7.0 just for experimental purposes. It is probably an insult to Taiyakixz since he is meticulous in his work, what I have done to the code, but I am anxious to move forward and he appears to have doubts, as well as being tied up with his personal affairs for a little while.
The attached ea allows for setting dss to false which is necessary for the inverse. Limits are treated as losses and stops are treated as profit targets.

This is just one application that has come to my attention and requires a bit of recoding to accommodate for the inverse. Perhaps a separate ea altogether, since many of the features of the original ea are not necessary for this simplified inverse version.
There are other applications that will become apparent as we explore the possibilities.
The basic strategy is to trade price action in relation to a hi lo channel.
The variations of this theme are what is being studied.
If it was so easy everyone would already be a millionaire. The truth is that most people become frustrated and quit. Those that are successful persevere.
I agree we do not want to simply go on "wild goose chases".

This history of trades taken by the ea in inverse shows the success.
With that I rest my case.

Mr. Taiyakixz to put his mind at ease.

But to remind that this inverse series is just one series. We will be returning to explore the use of the full service ea.

I made the ea(inverse version) and sent it to jiva.
Backtest results is below.

2010.01.04 - 2010.07.13 eurusd 5m
t/p=15, s/l=15


jiva will change the strategy again....?

Thank you, Taiyakixz.
We will be testing the ea.
again here is another history from just a few hours using inverse settings on 7.0 modified.

Taking the opposite position is obviously very consistent. As can be seen from the history attached.
If every time I take the opposite position for each of the trades listed that is very good profit.

So now we shall see how trading inverse signals compares to taking an opposite position.

Interesting studying the inverse..
Some interesting results for very simple ea.
no trailing stops, no trend moving averages, no shifts, no multi time frames, no nothing!!!
just reverse trade a break of the hi/lo channel

SymbolEURUSD (Euro vs. United States Dollar) Period5 Minutes (M5) 2010.01.04 02:00 - 2010.07.26 23:55 (2010.01.01 - 2010.07.27) ModelEvery tick (the most precise method based on all available least timeframes) ParametersiMagicNumber=362534101; Lots=1; TakeProfit=100; StopLoss=50; PendingPips=0; ExpireCandles=0; Channel_EMAPeriod=2; Channel_EMAShift=0; ShowComment=true;
Bars in test27707Ticks modelled3389482Modelling qualityn/a

Maybe Taiyakixz is right.
http://t0.gstatic.com/images?q=tbn:A...mQ2OtsHAt2_Hs=

The spread causes too much of a problem for the scalping. It was working nicely for some time but then the bad trades were coming.
The whole trade is shifted not just the cost of the spread when entering but also the difference between bid and ask when exiting.
It is not just a clear clean inverse, it is skewed.
So I tried.
I am glad the inverse ea was a simple thing. And not so complicated. Who knows there may be some good application for it down the road.
So back to v7.2 with so many nice features to try to capture the gains.
But v7.2 is not running properly.
In the tester it only takes few trades and then stalls out. Won't go any further. I have tried it on several platforms from different brokers. v7.2 is stalling out.
So I have gone back tov5.0 in order to just keep working. Until v7.2 is up and running.

Attached is the simple inverse ea just in case anyone wants to experiment with it...

so here is good results for EURUSD 4hr with a reasonable stop loss amount v5.0

SymbolEURUSD (Euro vs. United States Dollar) Period4 Hours (H4) 2010.01.04 00:00 - 2010.07.26 20:00 (2010.01.01 - 2010.07.27) ModelEvery tick (the most precise method based on all available least timeframes) ParametersiMagicNumber=8435501; GMTOffset=3; TradeStartHour=1; TradeEndHour=24; Lots=1; TakeProfit1=20; TrailingStop1=80; TakeProfit2=1000; TrailingStop2=100; StopLoss=80; PendingPips=5; ExpireCandles=1; DelayOrder2=0; CheckSlope=true; DSS_name="Robby DSS Bressert Colored with alert v1.1"; DSS_EMA=8; DSS_stoch=13; Use_T3MA=false; T3MA_name="T3MA"; T3MA_Periods=8; T3MA_VolumeFactor=0.7; Channel_EMAPeriod=38; Channel_EMAShift=0; ShowComment=true; EXP_MA_1="-------- MA settings --------"; EXP_MA_2="-- Method 0:SMA, 1:EMA, 2:SMMA, 3:LWMA"; EXP_MA_3="-- Timeframe 0:current or minutes"; Use_MA1=false; MA1_Timeframe=0; MA1_Period=0; MA1_Method=1; MA1_Shift=0; Use_MA2=false; MA2_Timeframe=0; MA2_Period=0; MA2_Method=1; MA2_Shift=0; Use_MA3=false; MA3_Timeframe=0; MA3_Period=0; MA3_Method=1; MA3_Shift=0; Use_MA4=false; MA4_Timeframe=0; MA4_Period=55; MA4_Method=1; MA4_Shift=0; Use_MA5=false; MA5_Timeframe=0; MA5_Period=197; MA5_Method=0; MA5_Shift=0;
Bars in test1574Ticks modelled3330281Modelling qualityn/a

These are 3 settings I have set up live on one account. They have been backtested only. Live forward testing is now in progess. Two trades for EURCHF were completed last week with 2 positions each.
They have been optimized. They can trade 2 or 3 times/month each. So look for 6 to 9 trades per month all 3 combined. They can and do catch big moves.
We will continue to search for more of the same type of settings. Would like to set up at least 3 more like these. More if possible.

So far AUDUSD, EURUSD, EURCHF....
similar settings just slightly tweaked for each.

Here are the specs:
using v5.0 FXDD

AUDUSD
iMagicNumber=6968560; GMTOffset=3; TradeStartHour=1; TradeEndHour=24; Lots=1; TakeProfit1=15; TrailingStop1=140; TakeProfit2=1000; TrailingStop2=200; StopLoss=140; PendingPips=55; ExpireCandles=4; DelayOrder2=0; CheckSlope=false; DSS_name="Robby DSS Bressert Colored with alert v1.1"; DSS_EMA=8; DSS_stoch=13; Use_T3MA=true; T3MA_name="T3MA"; T3MA_Periods=400; T3MA_VolumeFactor=0.7; Channel_EMAPeriod=2; Channel_EMAShift=0; ShowComment=true; no moving averages used...

EURUSD
iMagicNumber=6968560; GMTOffset=3; TradeStartHour=1; TradeEndHour=24; Lots=1; TakeProfit1=15; TrailingStop1=140; TakeProfit2=1000; TrailingStop2=200; StopLoss=140; PendingPips=55; ExpireCandles=6; DelayOrder2=0; CheckSlope=false; DSS_name="Robby DSS Bressert Colored with alert v1.1"; DSS_EMA=8; DSS_stoch=13; Use_T3MA=true; T3MA_name="T3MA"; T3MA_Periods=400; T3MA_VolumeFactor=0.7; Channel_EMAPeriod=2; Channel_EMAShift=0; ShowComment=true; no moving averages used

EURCHF
iMagicNumber=532753; GMTOffset=3; TradeStartHour=1; TradeEndHour=24; Lots=1; TakeProfit1=15; TrailingStop1=140; TakeProfit2=1000; TrailingStop2=200; StopLoss=140; PendingPips=25; ExpireCandles=1; DelayOrder2=0; CheckSlope=false; DSS_name="Robby DSS Bressert Colored with alert v1.1"; DSS_EMA=8; DSS_stoch=13; Use_T3MA=true; T3MA_name="T3MA"; T3MA_Periods=400; T3MA_VolumeFactor=0.7; Channel_EMAPeriod=2; Channel_EMAShift=0; ShowComment=true; no moving averages used

AUDUSD looks to be setting for the robot to trade short...

A sell stop has been triggered, both for AUDUSD and EURCHF

Worked all day to find setting for GBPJPY. No luck. Nothing consistent.
Threw the towel in...
Put the tester on USDJPY and wow. There it was.
So here are some sweet settings for USDJPY...
I'm adding this to my arsenal.

USDJPY 30 min...v5.0

iMagicNumber=532753; GMTOffset=3; TradeStartHour=1; TradeEndHour=24; Lots=1; TakeProfit1=15; TrailingStop1=140; TakeProfit2=30; TrailingStop2=200; StopLoss=140; PendingPips=36; ExpireCandles=3; DelayOrder2=0; CheckSlope=false; DSS_name="Robby DSS Bressert Colored with alert v1.1"; DSS_EMA=8; DSS_stoch=13; Use_T3MA=true; T3MA_name="T3MA"; T3MA_Periods=400; T3MA_VolumeFactor=0.7; Channel_EMAPeriod=55; Channel_EMAShift=0; ShowComment=true; no moving averages used