Simbasystem-gbpusd - page 45
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
Hey, thought I would stop in and give what I know about xpMA's.
A step period of 3-4 is best, 3 with a confirm on close to be precise, and that seems to be any type of MA.
SATL dont work well, at least in my preliminary testing so I moved on. It does merit further study, but there isnt much you can do with it.
JMA phase 100 is better, dont know why but it is.
My unfinished work led me to a 50period hull for the one hour charts. I had it making about 3000 pips a year,a pair, which aint bad for one moving average.
I have hundreds of year long backtests on moving averages, and still wasnt close to finding the right combination. I was looking for some combination to survive the aug-sep 2006 market and do well otherwise. So I started down the road of testing every period from 5-100 in 5 period increments with the intention to find the best reacting MA with no stops or profit targets, just the MA doing all the work. There are three distinct patterns with all of them. There is a short term, medium term, and long term period with each MA that works, with crap between them. Example.
A 10-15 period
A 45-50 period
A 70-75 period
Around those periods are most profitable while in between fails. I am sure there would be another level above that but just never got around to it. Each MA has its own unique and distinct characteristics. While one was good at reaction another was good at smoothing. You dont want a fast reaction on just a spike but you dont want smooth on a quick reversal. What im getting at is i feel there is a key out there to a combination of MA's. I personally think it is picking a period of time, fast, medium, or slow, and combining 3 different MA types. When the last one confirms, get in, when the first one signals out, get out. To find that combination is literally a thousand hours backtesting of which I dont have. I was hoping more people would have been interested in the MultixpMA as 10 people backtesting could get the job done.
Enough of my rambles. I am going to send Simba the source for multi and he can add his filters to it. I am also going to post another EA that is for a single MA that is more efficient for testing a single MA. Initially I would look to a combination of a Hull, and JMA. Beyond that i have no idea.
Good luck!
Hi all, I tried to repeat the first Simba's backtest but with no success. I do have the Alpari data with 90% quality, I do have the testing EA (and it works with default csv file) but when I enter my values into csv, it doesn't perform any trades. No errors released. I attach my EA settings file and the csv file, please tell me where I'm wrong. Maybe I just didn't see something, that would be no surprise, I feel sick again
Set your Sl and TP to 999 rather than 0. Check your journal tab in the backtester and see if it says something about invalid stops or something like that. if it does, thats it. Also attached is the doc for multixpma. And like I said, for a single MA, try the EA above, more efficient.
Thanks C.E.O
Hey, thought I would stop in and give what I know about xpMA's.
A step period of 3-4 is best, 3 with a confirm on close to be precise, and that seems to be any type of MA.
SATL dont work well, at least in my preliminary testing so I moved on. It does merit further study, but there isnt much you can do with it.
JMA phase 100 is better, dont know why but it is.
My unfinished work led me to a 50period hull for the one hour charts. I had it making about 3000 pips a year,a pair, which aint bad for one moving average.
I have hundreds of year long backtests on moving averages, and still wasnt close to finding the right combination. I was looking for some combination to survive the aug-sep 2006 market and do well otherwise. So I started down the road of testing every period from 5-100 in 5 period increments with the intention to find the best reacting MA with no stops or profit targets, just the MA doing all the work. There are three distinct patterns with all of them. There is a short term, medium term, and long term period with each MA that works, with crap between them. Example.
A 10-15 period
A 45-50 period
A 70-75 period
Around those periods are most profitable while in between fails. I am sure there would be another level above that but just never got around to it. Each MA has its own unique and distinct characteristics. While one was good at reaction another was good at smoothing. You dont want a fast reaction on just a spike but you dont want smooth on a quick reversal. What im getting at is i feel there is a key out there to a combination of MA's. I personally think it is picking a period of time, fast, medium, or slow, and combining 3 different MA types. When the last one confirms, get in, when the first one signals out, get out. To find that combination is literally a thousand hours backtesting of which I dont have. I was hoping more people would have been interested in the MultixpMA as 10 people backtesting could get the job done.
Enough of my rambles. I am going to send Simba the source for multi and he can add his filters to it. I am also going to post another EA that is for a single MA that is more efficient for testing a single MA. Initially I would look to a combination of a Hull, and JMA. Beyond that i have no idea.
Good luck!Thanks a lot C.E.O for your informative post...
I do not have a doubt about Hull ma,since that is what I used before finding SATL,with 45 periods,in 4h tf,and step 1..and it worked fine,so,probably your tested settings with 50 and step 3 on close are one of the best options..
Regarding SATL and JMA,I will check and learn and will probably arrive at the same conclusion as you,though I would like to know why Satl has given me such good results
In addition I have created several digital custom filters(SimbaSatl,SimbaFatl..) which are more responsive than Satl,while keeping you on trend..we will test them once finished with the 2 "best options" JMA(phase 100 and phase 0) and HULL ma and also the option which is working for me now,the SATL..
I will keep you informed of any new development..
Thanks again and Regards.
Simba
Malcik
Hi all, I tried to repeat the first Simba's backtest but with no success. I do have the Alpari data with 90% quality, I do have the testing EA (and it works with default csv file) but when I enter my values into csv, it doesn't perform any trades. No errors released. I attach my EA settings file and the csv file, please tell me where I'm wrong. Maybe I just didn't see something, that would be no surprise, I feel sick again
Hi Malcik..
You really should slow down on the Pivo
..jokes aside I hope you improve your condition.
This what I have in the xpma.csv when I want to test slope of JMA 13,phase 0..on 4h charts..you just put it and save as file xpma.csv
0,13,7,0,0.8,0,1,0,1,0
If you want to test the SATL..delete the above and input the following
0,13,10,0,0.8,0,1,0,1,0.........in this case the 13 is irrelevant,since it doesn`t affect the digital filter
If,for example you want to test JMA 13 on 4h charts for D1+4H slope agreement what you should input in the csv is
0,13,7,0,0.8,0,1,0,1,0
1440,13,7,0,0.8,0,1,0,1,0
Or if you want to test the crossover of the 2(the second crosses on the first..)
1440,13,7,0,0.8,0,1,0,2,0
0,13,7,0,0.8,0,1,0,2,0
And my settings for the EA are...
positions=2
deposit=50000
currency=USD
fitnes=0
genetic=1
xpmafile=xpma.csv
Order_Setting=---------- Order Setting
OneEntryPerBar=0
ReverseTradesOnly=0
StopLoss=1000.00000000
StopLoss,F=1
StopLoss,1=1000.00000000
StopLoss,2=10.00000000
StopLoss,3=1010.00000000
TakeProfit=1000.00000000
TakeProfit,F=1
TakeProfit,1=1000.00000000
TakeProfit,2=10.00000000
TakeProfit,3=1010.00000000
BreakEvenAfterPips=40
BreakEvenAfterPips,F=0
BreakEvenAfterPips,1=40
BreakEvenAfterPips,2=0
BreakEvenAfterPips,3=0
Slippage=5
Slippage,F=0
Slippage,1=5
Slippage,2=0
Slippage,3=0
MaxOpenTrade=1
MaxOpenTrade,F=1
MaxOpenTrade,1=1
MaxOpenTrade,2=1
MaxOpenTrade,3=6
MinPriceDistance=5
MinPriceDistance,F=0
MinPriceDistance,1=5
MinPriceDistance,2=0
MinPriceDistance,3=0
TrailingStop_Setting=---------- Trail Management Settings
TrailingStopType=1
TrailingStopType,F=0
TrailingStopType,1=1
TrailingStopType,2=0
TrailingStopType,3=0
TrailingStop=0
TrailingStop,F=0
TrailingStop,1=0
TrailingStop,2=0
TrailingStop,3=0
Moving_Average_Setting=---------- MA close Settings
CloseMA=0
CloseMA,F=1
CloseMA,1=0
CloseMA,2=1000
CloseMA,3=5000
RSI_Settings=---------- RSI Filter Settings
UseRSIfilter=0
RSIPeriod=21
RSIPeriod,F=0
RSIPeriod,1=21
RSIPeriod,2=0
RSIPeriod,3=0
RSIPrice=1
RSIPrice,F=0
RSIPrice,1=1
RSIPrice,2=0
RSIPrice,3=0
RSIBUYthreshold=50.00000000
RSIBUYthreshold,F=0
RSIBUYthreshold,1=50.00000000
RSIBUYthreshold,2=0.00000000
RSIBUYthreshold,3=0.00000000
RSISELLthreshold=50.00000000
RSISELLthreshold,F=0
RSISELLthreshold,1=50.00000000
RSISELLthreshold,2=0.00000000
RSISELLthreshold,3=0.00000000
CCI_Settings=---------- CCI Filter Settings
UseCCIfilter=0
CCIPeriod=21
CCIPeriod,F=0
CCIPeriod,1=21
CCIPeriod,2=0
CCIPeriod,3=0
CCIPrice=1
CCIPrice,F=0
CCIPrice,1=1
CCIPrice,2=0
CCIPrice,3=0
CCIBUYthreshold=50.00000000
CCIBUYthreshold,F=0
CCIBUYthreshold,1=50.00000000
CCIBUYthreshold,2=0.00000000
CCIBUYthreshold,3=0.00000000
CCISELLthreshold=-50.00000000
CCISELLthreshold,F=0
CCISELLthreshold,1=-50.00000000
CCISELLthreshold,2=0.00000000
CCISELLthreshold,3=0.00000000
Volume_Settings=---------- Volume Filter Settings
UseVolumeFilter=0
VolumeRange=1
VolumeRange,F=0
VolumeRange,1=1
VolumeRange,2=0
VolumeRange,3=0
VolumeThreshold=7.00000000
VolumeThreshold,F=0
VolumeThreshold,1=7.00000000
VolumeThreshold,2=0.00000000
VolumeThreshold,3=0.00000000
Time_Parameters=---------- EA Active Time
UseHourTrade=0
StartHour=0
StartHour,F=0
StartHour,1=10
StartHour,2=1
StartHour,3=15
EndHour=11
EndHour,F=0
EndHour,1=11
EndHour,2=0
EndHour,3=0
MM_Parameters=---------- Money Management
Lots=5.00000000
Lots,F=1
Lots,1=5.00000000
Lots,2=1.00000000
Lots,3=5.00000000
MM=0
AccountIsMicro=0
DecreaseFactor=0.00000000
DecreaseFactor,F=0
DecreaseFactor,1=0.00000000
DecreaseFactor,2=100.00000000
DecreaseFactor,3=500.00000000
Risk=10
Risk,F=0
Risk,1=10
Risk,2=0
Risk,3=0
Alert_Setting=---------- Alert Setting
EnableAlert=1
EmailAlert=0
SoundFilename=alert.wav
Testing_Parameters=---------- Back Test Parameter
PrintControl=1
Show_Settings=1
balance_enable=0
balance=200.00
profit_enable=0
profit=10000.00
marginlevel_enable=0
marginlevel=30.00
maxdrawdown_enable=0
maxdrawdown=70.00
consecloss_enable=0
consecloss=5000.00
conseclossdeals_enable=0
conseclossdeals=10.00
consecwin_enable=0
consecwin=10000.00
consecwindeals_enable=0
consecwindeals=30.00
Hope it is clear now..
Simba
Improved Testing Jma 13 4h
Hi All,
I DID ALL THE PROCESSES TO OBTAIN BETTER DATA QUALITY,AND,FOR THE MOMENT THIS IS THE BEST iIHAVE..76%
Test Results for 4h GBPUSD JMA 13,phase 0..interesting PF=2.28
Bars in test 671
Ticks modelled 494948
Modelling quality 76.60%
Initial deposit 50000.00
Total net profit 83245.38
Gross profit 148515.38
Gross loss -65270.00
Profit factor 2.28
Expected payoff 1172.47
Absolute drawdown 0.00
Maximal drawdown 13409.95 (16.22%)
Relative drawdown 16.22% (13409.95)
Total trades 71
Short positions (won %) 35 (51.43%)
Long positions (won %) 36 (44.44%)
Profit trades (% of total) 34 (47.89%)
Loss trades (% of total) 37 (52.11%)
Largest
profit trade 27200.01
loss trade -6799.97
Average
profit trade 4368.10
loss trade -1764.05
Maximum
consecutive wins (profit in money) 7 (23870.09)
consecutive losses (loss in money) 5 (-8610.08)
Maximal
consecutive profit (count of wins) 27200.01 (1)
consecutive loss (count of losses) -9205.09 (4)
Average
consecutive wins 2
consecutive losses 2
Please see pic
Simba
Testing 1h JMA 13
Hi All,
Same test as previous post ,86% modelling quality ,but on 1h tf and 1h JMA 13..it doesn`t work..4h is much better
Bars in test 2655
Ticks modelled 570024
Modelling quality 86.61%
Initial deposit 50000.00
Total net profit -41069.77
Gross profit 91040.19
Gross loss -132109.96
Profit factor 0.69
Expected payoff -285.21
Absolute drawdown 41069.77
Maximal drawdown 47399.62 (84.15%)
Relative drawdown 84.15% (47399.62)
Total trades 144
Short positions (won %) 72 (27.78%)
Long positions (won %) 72 (31.94%)
Profit trades (% of total) 43 (29.86%)
Loss trades (% of total) 101 (70.14%)
Largest
profit trade 6494.97
loss trade -5350.03
Average
profit trade 2117.21
loss trade -1308.02
Maximum
consecutive wins (profit in money) 3 (5535.04)
consecutive losses (loss in money) 11 (-14425.23)
Maximal
consecutive profit (count of wins) 6494.97 (1)
consecutive loss (count of losses) -15459.95 (8)
Average
consecutive wins 1
consecutive losses 3
Please see pic
Simba
4h Jma 13 Phase 100
Hi All,
Same as 2 posts before JMA 13 ,4H PHASE 100..results are good but slightly worse than for PHASE 0..PF=1.89
Bars in test 671
Ticks modelled 494948
Modelling quality 76.60%
Initial deposit 50000.00
Total net profit 69380.35
Gross profit 147425.31
Gross loss -78044.96
Profit factor 1.89
Expected payoff 816.24
Absolute drawdown 0.00
Maximal drawdown 15014.78 (15.35%)
Relative drawdown 15.35% (15014.78)
Total trades 85
Short positions (won %) 40 (47.50%)
Long positions (won %) 45 (46.67%)
Profit trades (% of total) 40 (47.06%)
Loss trades (% of total) 45 (52.94%)
Largest
profit trade 15399.98
loss trade -8599.97
Average
profit trade 3685.63
loss trade -1734.33
Maximum
consecutive wins (profit in money) 5 (7825.18)
consecutive losses (loss in money) 5 (-8960.08)
Maximal
consecutive profit (count of wins) 22139.93 (2)
consecutive loss (count of losses) -8960.08 (5)
Average
consecutive wins 2
consecutive losses 2
Please see pic
Test Jma 13 4h Phase -100
Test results for phase -100,while still good,worse than phase 100 and phase 0
Bars in test 671
Ticks modelled 494948
Modelling quality 76.60%
Initial deposit 50000.00
Total net profit 45335.27
Gross profit 123575.13
Gross loss -78239.86
Profit factor 1.58
Expected payoff 731.21
Absolute drawdown 0.00
Maximal drawdown 17744.98 (25.23%)
Relative drawdown 25.23% (17744.98)
Total trades 62
Short positions (won %) 28 (42.86%)
Long positions (won %) 34 (32.35%)
Profit trades (% of total) 23 (37.10%)
Loss trades (% of total) 39 (62.90%)
Largest
profit trade 27334.98
loss trade -6650.00
Average
profit trade 5372.83
loss trade -2006.15
Maximum
consecutive wins (profit in money) 5 (23365.09)
consecutive losses (loss in money) 4 (-13160.07)
Maximal
consecutive profit (count of wins) 27334.98 (1)
consecutive loss (count of losses) -13160.07 (4)
Average
consecutive wins 1
consecutive losses 2
Simba
Test Satl 4h
Hi All,
Test results on SATL 4H for GBPUSD..Positive,PF=1.43..but worse than JMA 13...Well,if Malcik and Dabella ,who are also working on this , can replicate my results..I will be convinced(I practically am) to switch fromSATL to Jma..
In the meantime ,I am going to test for 1h SATL,4H SIMBA SATL ,and different periods for JMA..15,19,25,39,50,53 ON 4H TF..I will post the conclusions to avoid boring everyone with repetitive details...
Bars in test 671
Ticks modelled 494948
Modelling quality 76.60%
Initial deposit 50000.00
Total net profit 34445.03
Gross profit 113955.06
Gross loss -79510.03
Profit factor 1.43
Expected payoff 538.20
Absolute drawdown 3499.98
Maximal drawdown 24100.08 (22.20%)
Relative drawdown 23.39% (18179.95)
Total trades 64
Short positions (won %) 31 (38.71%)
Long positions (won %) 33 (18.18%)
Profit trades (% of total) 18 (28.13%)
Loss trades (% of total) 46 (71.88%)
Largest
profit trade 39319.94
loss trade -6499.98
Average
profit trade 6330.84
loss trade -1728.48
Maximum
consecutive wins (profit in money) 3 (42969.94)
consecutive losses (loss in money) 13 (-21235.05)
Maximal
consecutive profit (count of wins) 42969.94 (3)
consecutive loss (count of losses) -24100.08 (8)
Average
consecutive wins 2
consecutive losses 4
Simba
Hi all,
I finally seem to have solved the technical problems. The reason the EA didn't work with other than factory csv is that I forgot to put an empty line at the end of the csv.
I have first 90% quality backtest for you - it's JMA 13 phase 0 on H4. Everything is working fine, I randomly checked some of its entries and they are correct. Still I'm not really sure about it though - in my backtest, it basically wiped out the 50 grant account in 5 months which doesn't look real given that it was profitable for Simba at 70%-80% quality.
If someone else gets 90% quality, please do this backtest too so that we can compare before I move on with backtesting. Attached is the full report with all the trades and with equity curve. This report contains my EA setting as well -- take a look at those too. Second is the csv file.