Welcome to Forex-TSD!, one of the largest Forex forums worldwide, where you will be able to find the most complete and reliable Forex information imaginable.
From the list below, select the forum that you want to visit and register to post, as many times you want. It’s absolutely free. Click here for registering on Forex-TSD.
Exclusive Forum
The Exclusive Forum is the only paid section. Once you subscribe, you will get free access to real cutting-edge Trading Systems (automated and not), Indicators, Signals, Articles, etc., that will help and guide you, in ways that you could only imagine, with your Forex trading.
Elite Section
Get access to private discussions, specialized support, indicators and trading systems reported every week.
Advanced Elite Section
For professional traders, trading system developers and any other member who may need to use and/or convert, the most cutting-edge exclusive indicators and trading systems for MT4 and MT5.
i think i have caused a lot of misunderstanding here. I have to apologize to Simba and cementman. I got the system from a Chinese forum. Though he did not want to disclose his indicators, but i felt i read it somewhere else,so i come to forex-tsd.com to check again and I found out it is from Simba. I am very sorry, perhaps i should be quiet and just learn from you guys.
ok,i have a question, we should follow the rules, buy if Satl is blue and william is below -5, and vice versa. Should we consider chandelierstop for example right now gbp/usd, the Satl is red, but chandelier has changed from sell to buying. So which one we should follow, or we just stay out of the market for the moment? tq.
ok,i have a question, we should follow the rules, buy if Satl is blue and william is below -5, and vice versa. Should we consider chandelierstop for example right now gbp/usd, the Satl is red, but chandelier has changed from sell to buying. So which one we should follow, or we just stay out of the market for the moment? tq.
Hi yan,
You just stay out of the market,as per the system,what the indicators relative positions are telling you is that this is a period of no defined trend..
If you are very experienced and have other confirming indicators,AND THIS IS NOT THE SYSTEM,JUST SOME PERSONAL THOUGHTS,you can start prepositioning for a potential change of trend..usually the first step in a trend change is the price crossing the satl followed by chandelier changing..but to stay within the system,you need confirmation by satl..and a wpr entry
Regarding your other post I accept your apologies,though they were not necessary ..
I would like to work with optimization, but I did not understand well the explanation below:
To find the above mentioned results i tested,all possible chand distance combinations-for tp 1000- from 1/200,10/200....1/50...etc,then i tested tp strategies-for chand1/200 -from 10 to 100,in steps of 10,then from100 to 1000 in steps of 100...then i combined the best "zones" (chand 35/40 to 90/120..and tp from 200 to 400) to find ,first an optimum,and second ,if the improvement was "across the board",which is a sign of consistency, ie:if applying chand 35/110 to all tp strategies(with previous chand 1/200) improved all of them across the board(yes,it did ),or,if applying TP300,350,400 TO CHAND 35/90..40/120(with previous tp 1000) improved them across the board...(yes it did ,again )..Between 40 and 50 tests,but Ithink it was worth the time..
Please... can you give more details of as carries through optimization?
Thanks...
Angelo
Hi Angelo,
Basically what I did is testing the alternatives to ,first find the best option,and second,to see if it was consistent and rational(ie:imagine that we find that the best option is chand distance 0-1000,and Tp 100..but for tp from 90 to 99 and from 101 to 200 the system loses money..something is clearly not working here,probably overfitting..hope i made it clear)
1-If we want to test tp from 10 to 100,steps of 10(10 tests) plus TP from 200 to 1000(9 tests more) we have a subtotal of 19 tests.
2-If we want to test for chandelier distances..10/200..in steps of 10..we have 19 tests going down(10/200,10/190..10/20) multiplied by18 going up (20/200,20/190..20/30)multiplied by 17 going up(30/200..)...so,basically we have 19 "factorial"(i don`t know the name in english..it is 19*18*17*...) tests..
3-now we have 19 combinations from point1 and approx 19¡(factorial) for point 2..that we have to multiply..solutions:1-brute force computer..not applicable..at least in my case
2-genetic optimizer..i did..tonight,the optimum it found was CHAND 35/105 with Tp 358 pips total profit 44.8k(so sligthly more than i found) basically the same pf and the same max%dd..it took all the night and i have the doubt that it tested only between 35/105(37/100,etc but not 10/200 ,25/120,etc..outside the boundaries..) for Chand distance,to find the best tp strategy..i will try to redo tonight or tomorrow night..
3-heuristic search,just divide the search space into "logical chunks" for the different testable factors,test each one independent of the other(this is why i initially tested tp strategies10,20,..1000 for chandelier distance 1/200..),look for groups of optimum and semioptimum zones(chandelier 35/40 in the minimum 90/120 in the maximum..tp from 200 to 400 pips),and then reduce the "combination search" to this reduced space..this is what i did,and the results were very near the optimum..
If you have enough free computer power ,an interesting thing to do is to click the genetic optimizer,with the widest settings..like chand 1/200 tp 1000..it will need from10 to 60 hours to do the search...if you don`t have this expendable comp power.you can reduce the search,for example to the low profit alternatives(ie tp 30/70) with wide chand distance...to see if there is an "anxiety free" quick bucks alternative ...
again you reply my post so fast. i got what you mean, now i am looking into your system, and god, how can i missed out this good system. i need to work harder to compensate what i have lost. great work Simba, u are 45 now right, from the way you answer and handle things, you are a good man. thanks again.
I did the last backtest. The results are slightly worse -- I guess it's just a slight difference of the time period. For some strange reason I can't save them as report so I'm just providing basic ratios and resulting equity curve.
Init. deposit: $10,000
Net profit: $41,856.50
PF: 2.00
Expected payoff: 176.61
Win ratio: 35.44%
Max. DD: $3,963 (13.24% of the account at the moment; 39.63% of the initial account)
__________________
Malcik
Last edited by Malcik; 03-07-2007 at 09:08 AM.
Reason: Found a way to copy the results into a text file, please check :)
Hey Simba, that's a good news. Another good news is that with your data files I now have 90% quality I tried to redo this latest backtest but it's taking too long, I want to go to sleep so I stopped it. I'll redo all of them tomorrow and if there is no major difference, I'll also post new version of the EA that will be able to test W%R settings.
As far as the contest, I don't really care. I don't know anything about contests and what they are good for; please tell me the advantages for me and for us all
Anyway, the system is Simba's, the EA is just a tool to trade it and I don't think I'm the only one capable of coding it. So if you can't name any advantage for us generally but you have one for yourself, that's fine with me too -- go ahead as long as it's fine with Simba.
See you tomorrow
Hi Malcik ,
Glad to hear you can start doing the tests with the data i sent you,thanks for wpr distances option..and hope you slept like a baby
Contest:
1-Advantages for you:they are good for your ego(if you win..self explanatory..if you lose..nothing for the ego like a reality check) and for your pocket(if your system shows consistent results with low drawdowns,it is easier to interest external investors,they come to you..)..also contests "must be"(i presume,never been in one ) fun and a challenge,and ,obviously,you will learn from the experience
2-Advantages for "all us": if a public system "of " forex-tsd wins/finish well ,
A-it will be good for the forum,since the implication is that in this forum people know what are they talking about and can create profitable systems,this will probably increase the number of members expanding the learning options for everybody.
B-It will incentivate people to read and learn about our methodology(it is a manual trading system..after all ) learning the core root of trading(trade with the trend,buy dips in uptrend,sell rallies in downtrend),and probably it will help many newbies to delay they inevitable self destruction expanding their learning processes at the same time
C-Hopefully,by reading the thread and learning from our mistakes,newbies will understand the necessity of testing their assumptions,and that EAs have a function even in manual systems trading
3-Advantages for me..same as for you,but I won`t do it alone,if i do it ,it should be fun and challenging..in addition i don`t think our EA as it is,though good for trading(it is safe and profitable) ,can stand any chance in a contest where winners usually use some kind of selfadapting trend filters(some would call it overfitting..i am starting to think that this is the way to go in the future..) with very aggressive money management(that we still haven`t started to use.. )..basically:if you are interested,if somebody else-that can help us- wants to be included in our team,AND if we have enough time to,first improve the EA,and second to devise a special contest version..I am interested..
The first thing would be to check the timing for the contest..but first..
Glad to hear you can start doing the tests with the data i sent you,thanks for wpr distances option..and hope you slept like a baby
Contest:
1-Advantages for you:they are good for your ego(if you win..self explanatory..if you lose..nothing for the ego like a reality check) and for your pocket(if your system shows consistent results with low drawdowns,it is easier to interest external investors,they come to you..)..also contests "must be"(i presume,never been in one ) fun and a challenge,and ,obviously,you will learn from the experience
2-Advantages for "all us": if a public system "of " forex-tsd wins/finish well ,
A-it will be good for the forum,since the implication is that in this forum people know what are they talking about and can create profitable systems,this will probably increase the number of members expanding the learning options for everybody.
B-It will incentivate people to read and learn about our methodology(it is a manual trading system..after all ) learning the core root of trading(trade with the trend,buy dips in uptrend,sell rallies in downtrend),and probably it will help many newbies to delay they inevitable self destruction expanding their learning processes at the same time
C-Hopefully,by reading the thread and learning from our mistakes,newbies will understand the necessity of testing their assumptions,and that EAs have a function even in manual systems trading
3-Advantages for me..same as for you,but I won`t do it alone,if i do it ,it should be fun and challenging..in addition i don`t think our EA as it is,though good for trading(it is safe and profitable) ,can stand any chance in a contest where winners usually use some kind of selfadapting trend filters(some would call it overfitting..i am starting to think that this is the way to go in the future..) with very aggressive money management(that we still haven`t started to use.. )..basically:if you are interested,if somebody else-that can help us- wants to be included in our team,AND if we have enough time to,first improve the EA,and second to devise a special contest version..I am interested..
The first thing would be to check the timing for the contest..but first..
..tell me ,tell me..yes or not
Regards
Simba
Well I got a great 5-hour sleep -- one so short that had I been doing it like that when I was a baby, I wouldn't be around today
OK, well it looks interesting; I don't want to do it alone but if we all could participate I think it could be useful... and if not, than just for fun.
It's true that SimbaSystem is a conservative one and the winners of contests are very agressive, i.e. Phoenix and others. Anyway, why not participate
Still I guess we still need to work on our tests and optimization, and also on money management. Contest would then be good for forward testing
As far as the self-adapting systems: yes, I know about that idea and how large a potential there is. Do you have some knowledge and experience about this? Don't the digital filters have a form of this coded in them? I've seen such terms in Rocket Science for Traders... am I right? Frankly, I'm having a hard time understanding the book. I thought I was a science guy, well I still think I am but not as good as I thought I understand the idea and most of what he's doing but I don't have a clue about how to use the knowledge in my own trading. Maybe John Ehlers just assumed that his readers have some previous knowledge about digital filters. Isn't there anything I should have read before reading this book?
Well I got a great 5-hour sleep -- one so short that had I been doing it like that when I was a baby, I wouldn't be around today
OK, well it looks interesting; I don't want to do it alone but if we all could participate I think it could be useful... and if not, than just for fun.
It's true that SimbaSystem is a conservative one and the winners of contests are very agressive, i.e. Phoenix and others. Anyway, why not participate
Still I guess we still need to work on our tests and optimization, and also on money management. Contest would then be good for forward testing
As far as the self-adapting systems: yes, I know about that idea and how large a potential there is. Do you have some knowledge and experience about this? Don't the digital filters have a form of this coded in them? I've seen such terms in Rocket Science for Traders... am I right? Frankly, I'm having a hard time understanding the book. I thought I was a science guy, well I still think I am but not as good as I thought I understand the idea and most of what he's doing but I don't have a clue about how to use the knowledge in my own trading. Maybe John Ehlers just assumed that his readers have some previous knowledge about digital filters. Isn't there anything I should have read before reading this book?
Hi Malcik,
Well,5 hours sleep gives you more time for real life,so you should be grateful for it..
Ok,let`s check the dates for the contest,etc,etc
Ehlers book..well,you could try with the summer edition of Penthouse..it will put you off the digital filters mode but i am sure you will understand it much better ..jokes aside,just get the concepts,he already did the math for you..and,if you can,go visit Damiani site,this guy has a fantastic understanding about applying ehlers indicators to trading..in addition,there is a ehlers indicators thread in "indicators" in this forum..there are many ehlers indicators posted there,most by bluto..this should be enough to get you busy until your next 5 hours baby sleep..or Penthouse reading
Self adaptive:yes,most digital filters self adapt,what is different for the contest versions i referred to is that they optimize the settings every2 /4 weeks..basically,if the old settings still work,they leave them,if they start decreasing performance,they substitute for new ones..
Money management :next step should be to start thinking about it and include it in the EA..if you want,do a search in the internet for "fixed ratio" money management,IMHO this is one of the strategies we should try to test,once we have a nice system,this kind of mm can improve the results spectacularly,and,it can also improve the total results for high%pft variations(like tp 35 pips,etc)making them very very workable..
I leave you,cable is starting to be where i want it...and i have to trade
During sleepeng time my computer did optimization for my data stream. For 5 hours was 150 passes.I Could not get 44k, best result - 20k.I feel little envie to you,guy's, because seems like my data provider don't want for me to use SimbaSystem . Any way, I will do more optimization on weekend.
By the way, I red Hurst "Profit Magic", half book for the day.Thank you,Simba, for such nice recomendation.
Question:
Do you consider GBP/USD as the best pair for cycle analysis, or there some other cyclycally consistent pair's?
Best regards.
Lucky you to live in warm climate, we have today -20 Celsium outside.
I am very sorry, perhaps i should be quiet and just learn from you guys.
I tried to redo this latest backtest but it's taking too long, I want to go to sleep so I stopped it. I'll redo all of them tomorrow and if there is no major difference, I'll also post new version of the EA that will be able to test W%R settings.
Lucky you to live in warm climate, we have today -20 Celsium outside.
Linear Mode
