Simbasystem-gbpusd - page 89
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
And with MM
Grrrr, ZigZag Equity.... if EA have max 100lots order, then stop MM. Still VERY BIG DD.
What add FX Sniper Ergodic for confirmation entry after %R signal?
I used Google translator to translate the previously posted link and laughed my ass off! It's in Borat-speak!
BTW, this is the kind of stuff that got me banned from FF. They would categorize this as cultural insensitivity.
"Where prices are not such as to be?" That's funny, I don't care who you are. High five!
Sorry I've been away a little, but I now have a complete set of entirely Alpari M1 data since 1/3/2001 for GBP/USD. This was a real pain in the ass but it worked out. Here's how to do it:
1. Go to http://forum.viac.ru/viewtopic.php?t=2420
2. Download their GBP/USD data. It's on the second page. I don't speak Russian, only English and German, so I have no idea what anything means on that site except for the fact that this is the Alpari data from 2001 to 2005, and I can see non-Cyrillic letters that denote GBP/USD.
3. Import that data into your testing version of MT4 per Hendrick, i.e., delete all .hst files, etc.
4. Also import, right after that, at the same time, your current Alpari data. MT4 will merge the two sets. There is no need to do all this exporting to wordpad and holding your finger on a mouse for 3 hours to scroll through 1 million lines of data.
Close MT4 and re-open, open an offline 1M chart and proceed to period convert the data.
I see this as being pretty damn alright since we now have a complete set of data that is just from Alpari that is of a respectable length to achieve some sort of statistical reliability while all being, allegedly, from the same source. Although I still don't trust the Russians...
Duke
BTW, I would like to participate in this next round of testing but I downloaded v1.4b and it won't take any trades. 1.3 official still works okay. Any suggestions?
I used Google translator to translate the previously posted link and laughed my ass off! It's in Borat-speak!
BTW, this is the kind of stuff that got me banned from FF. They would categorize this as cultural insensitivity.
"Where prices are not such as to be?" That's funny, I don't care who you are. High five!Don't belive in Google machine translation.
On the first post admin of that forum (AlexSilver) asked people:
indicators will be attaching for the long time on the intra-day
chart showing 5 year data.And he mentioned about some gaps in the data.
He said that it may be some gaps but Goodman
(the other admin of that forum) checked and described every single gap
in all the data posted.
In next post some user found some mistakes. But admin (AlexSilver) spoke with him and found that the user place this MT4 data to MT3 Metatrader.
Franta Tests
I testing with new data 1.3.official and 1.3.b
Here result.
Before 2004... bad performance, but maybe only bad data. I do not know.
I think, backtest before 2004 is not trustful.Hi Franta,
Tests are good you should have done the first one ,that you did,june 04 to feb07 and a second one NOT FOR THE FULL PERIOD,but for THE REST OF THE PERIOD..jan01 to mid june 04..please see attached..and read my explanation`s point 3
EXPLANATION:;)..1-We optimized a system on 2.7 years of data with very good results for that data..this doesn`t mean a scheisse,until we have OUT OF SAMPLE proof that the system works positively in data sets that haven`t been used to optimize it,since we are going to employ the system in the future,that obviously has not been seen by the "optimizer"...NOW WE DO..we have 2 proofs,first the system worked for the same period(june04 to feb07) on ANOTHER PAIR..GBPJPY..and second,and more important..it works in the same pair,FOR A DIFFERENT PERIOD..jan01 to midjune04..3.5 years of data
2-From now on,DO NOT MIX THE DATA SETS,we will use them separately..
3-Obviously we can not expect the system to work out of sample as well as in sample,we can just expect it to have a profitable behaviour,so,the fact that the results are not as positive as in sample is normal,and when we start the mm phase we have to use out of sample results to estimate the kelly ratio,the delta and the lot management..in order to optimize..IN REALITY..the future live results
Now what we can do is ..
1-Franta ,if you are there,test the official settings on GBPJPY JAN01 TO MIDJUNE04..and post the results,jointly with previously posted GBPJPY 04/07..if the system works out of sample in both another pair and another period(ie gbpjpy 01 to 04) we have something worth exploring
..if you are not there..;)..i will download,recompile the data for gbpjpy and do and post the test
2-Malcik:There is a problem with 1.4 version,both duke and myself have been unable to make it trade..please take a look and solve the problem,then we will start to test the pivots issue
3-Once we have data from 1 and 2 ,we will start the isue of improving the main aspect of the system which is entry and exit tuning..we have from 30 to 40% profitable trades only,yes the system is profitable because av win is much bigger than avloss..and it will be even more once we improve the winning ratio..Franta and Malcik have many ideas..like substituting pivots for murrey levels,camarilla..changing from wpr to waddah attar explosion,confirmation by fx sniper`s cci,etc..I have 2 basic ideas that are,first to create a basic exit system that combines the trend staying properties of chandelier with the profit saving capability of PSAR..and second,to use a volatility filter to improve the winning ratio of entries...We will extensively test all of these suggestions,and more to be suggested,BUT FIRSTwe have to be sure that we have a WORTHY TRADING SYSTEM BENCHMARK that works out of sample..then we will put more meat into it..and hopefully most of it will be muscle..;)..and we will try to do it without adding unnecessary fat..the simplest the better
Regards
Simba
Sorry I've been away a little, but I now have a complete set of entirely Alpari M1 data since 1/3/2001 for GBP/USD. This was a real pain in the ass but it worked out. Here's how to do it:
1. Go to http://forum.viac.ru/viewtopic.php?t=2420
2. Download their GBP/USD data. It's on the second page. I don't speak Russian, only English and German, so I have no idea what anything means on that site except for the fact that this is the Alpari data from 2001 to 2005, and I can see non-Cyrillic letters that denote GBP/USD.
3. Import that data into your testing version of MT4 per Hendrick, i.e., delete all .hst files, etc.
4. Also import, right after that, at the same time, your current Alpari data. MT4 will merge the two sets. There is no need to do all this exporting to wordpad and holding your finger on a mouse for 3 hours to scroll through 1 million lines of data.
Close MT4 and re-open, open an offline 1M chart and proceed to period convert the data.
I see this as being pretty damn alright since we now have a complete set of data that is just from Alpari that is of a respectable length to achieve some sort of statistical reliability while all being, allegedly, from the same source. Although I still don't trust the Russians...
Duke
BTW, I would like to participate in this next round of testing but I downloaded v1.4b and it won't take any trades. 1.3 official still works okay. Any suggestions?Hi dukeofdets,
Thanks for getting back to us. The tutorial I posted makes sure everything works out fine. Because for me, for example, it didn't. I thought the same like you did: I imported the old set of data, and right after that I imported the new one, thinking that the overlapping part of the old set would be overwritten by the new set... But in reality, nothing of the new set was really imported and the HST files remained filled with the first set only.
MT4 is a great software with virtually limitless possibilities... but in some issues (data is probably the most famous one) it has bugs. So I just wanted to post a tutorial that will work for sure, although it's a lot more complicated, I agre
Please, attach missing Dolly.ex4 or better mq4
Sorry, I forgot about that, here it is.
I used the Dolly pivots instead FibLines because the lines are identical and Dolly is more suitable for backtesting (it draws short lines for each day on chart whereas FibLines only draw lines for today). For EA backtesting it wouldn't matter - only for visual it would - but I just wanted to keep it unified.
It is also better in the manner that it doesn't consider last hour or two of Sunday GMT, when the markets are already open, a full trading day. This only applies if your broker uses timezone other than CET/CEST (Alpari does, most other brokers don't). So if you use Alpari data, use this Dolly, if something else, please change the line #97 so that it includes the condition that is commented out.
I'll let you know about the other issues.
Update:
Sorry guys, but it really works for me... here are the results from Jan 2007 till today:
1 2007.01.03 08:10 buy 1 1.00 1.9712 0.0000 2.0067 0.00 10000.00
2 2007.01.03 12:00 close 1 1.00 1.9628 0.0000 2.0067 -840.00 9160.00
3 2007.01.26 08:25 sell 2 1.00 1.9667 0.0000 1.9315 0.00 9160.00
4 2007.01.30 03:00 close 2 1.00 1.9644 0.0000 1.9315 223.60 9383.60
5 2007.02.08 08:31 buy 3 1.00 1.9685 0.0000 2.0040 0.00 9383.60
6 2007.02.08 10:00 close 3 1.00 1.9668 0.0000 2.0040 -170.00 9213.60
7 2007.03.08 00:24 buy 4 1.00 1.9278 0.0000 1.9633 0.00 9213.60
8 2007.03.12 14:00 close 4 1.00 1.9317 0.0000 1.9633 395.40 9609.00
9 2007.03.14 15:23 sell 5 1.00 1.9300 0.0000 1.8945 0.00 9609.00
10 2007.03.14 16:00 close 5 1.00 1.9335 0.0000 1.8945 -350.00 9259.00
11 2007.03.26 15:27 sell 6 1.00 1.9656 0.0000 1.9301 0.00 9259.00
12 2007.03.26 16:00 close 6 1.00 1.9676 0.0000 1.9301 -200.00 9059.00
The settings are:
UseSATLFilter=1
ChandDist_Min=0
ChandDist_Max=10000
UseChandIfTouched=0
WPR_Period=90
WPR_TopLevel=-5
WPR_BottomLevel=-95
NextPivotMaxDist=5
LastPivotMaxDist=5
PivotWaitingTime=30
TPPips=355
atr_for_tp=0
atr_period=14
atr_multiplier=100.00000000
AllowReentries=0
MaximumRisk=0
I once again attach the EA to make sure we are talking about the same thing (only change I've made to it is a minor technical error discovered by Franta). I guess it should all be caused by the fact that I forgot to attach the Dolly file.
If it still doesn't work, please check the Journal tab of your Strategy Tester for possible errors.
If it works, I'll proceed with further updates -- not until then, we have to first make it trade
Hi Franta
1-Your b variation has worked very well out of sample too,why don`t you test it too for gbpjpy,separately,both testing periods(01-04 and 04/07)?
2-Regarding your perception of the tests,if we take the full period,this sytem makes like 53/54k(depending on a or b variation) in 380 trades approx,since we are talking 10$ per pip for 1 lot per 10 k in account we are talking about 5300 pips divided ny 380 trades approx 13 or 14 NET pips per trade for a system that trades on average approx 5.2 trades per month..so,we have around 70 NET pips per month ie 700$NET per month for a 10k account,we are talking 7% monthly return,which we will first improve,and then compound with the adequate money management strategy..
3-Probably 7% per month might seem very low,when you see the graphs of some systems..do not believe them,these rags to richess systems are a joke..see attached,I can create them in 15 minutes,BUT THEY NEVER WORK OUT OF SAMPLE..I only include the .gif,since the concept it is based on is something that i use as the basis of some very private trading strategies that I don`t want to share ,for the moment,in public..you only have to take a valid concept,overoptimize it,and you have an exceptional system that doesn`t work out of sample..
4-If your personal,or other forum members,situation regarding starting capital,etc,makes you despair at the thought of having to compound from a low base and to resort to overoptimized startegies hoping that they work long enough for you...think twice about it..if you are able to show real money results,7% monthly with controlled dd would be considered exceptional,you are going to be offered to handle other people money,for a fee,and then you will be able to increase your networth the right way..see attachment for the bad way
and smile..it doesn`t work out of sample
Regards
SimbaThis your test Equity my dream!
Franta Test 2
next and next
Hi Franta
1-Your b variation has worked very well out of sample too,why don`t you test it too for gbpjpy,separately,both testing periods(01-04 and 04/07)?
2-Regarding your perception of the tests,if we take the full period,this sytem makes like 53/54k(depending on a or b variation) in 380 trades approx,since we are talking 10$ per pip for 1 lot per 10 k in account we are talking about 5300 pips divided ny 380 trades approx 13 or 14 NET pips per trade for a system that trades on average approx 5.2 trades per month..so,we have around 70 NET pips per month ie 700$NET per month for a 10k account,we are talking 7% monthly return,which we will first improve,and then compound with the adequate money management strategy..
3-Probably 7% per month might seem very low,when you see the graphs of some systems..do not believe them,these rags to richess systems are a joke..see attached,I can create them in 15 minutes,BUT THEY NEVER WORK OUT OF SAMPLE..I only include the .gif,since the concept it is based on is something that i use as the basis of some very private trading strategies that I don`t want to share ,for the moment,in public..you only have to take a valid concept,overoptimize it,and you have an exceptional system that doesn`t work out of sample..
4-If your personal,or other forum members,situation regarding starting capital,etc,makes you despair at the thought of having to compound from a low base and to resort to overoptimized startegies hoping that they work long enough for you...think twice about it..if you are able to show real money results for a prolongued time,5% monthly with controlled dd would be considered exceptional,you are going to be able to ask to handle other people money,for a fee,and then you will be able to increase your networth the right way,without unnecessary risks..see attachment for the bad way
and smile..it doesn`t work out of sample
Regards
Simba
trading the GBPJPY WITH SIMBA SYSTEM!
hi guys.
i have discover a indicator that can help us when to tack profits with the GBPJPY .
i took it from the XO METHOD .
i will attach images to explain , i'm steel try to back test it but it's look good.
i find that this could work only with this pair, i'ts not working 100% of the times but could be very helpull to know where the resistense or support are going to end.