Trading Strategies Based On Digital Filters - page 27
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
well...here's my latest...after 5 hours of working on this stuff tonight, i'm afraid i'm exhausted....
i feel like i'm close in understanding how to create these cycles....
the picture represents a 30 min at 21,33,36,39 ( great peak all the way to the top at 34) and a 15 min of 55,66,70,77 ( great peak at 68).
34x2 = 68. Both were well defined peaks. I just feel like i am missing something with regards to creating these. I'm looking for high, solo peaks, etc. Any advice would be much appreciated.
Dvarrin you still around?
cl
well...here's my latest...after 5 hours of working on this stuff tonight, i'm afraid i'm exhausted....
i feel like i'm close in understanding how to create these cycles....
the picture represents a 30 min at 21,33,36,39 ( great peak all the way to the top at 34) and a 15 min of 55,66,70,77 ( great peak at 68).
34x2 = 68. Both were well defined peaks. I just feel like i am missing something with regards to creating these. I'm looking for high, solo peaks, etc. Any advice would be much appreciated.
Dvarrin you still around?
clClahn,try to use 33,35 for a 34 cycle..same for 68 cycle..use 67,69.
My opinion..do the peak and valley multiple time frame analysis on full history,not on 200 bars..you are looking here for a cycle that is persistent..and look for a harmonic cycle between 2 harmonic timeframes..example..in h4,h1...a 15periods and 60 periods...do not go below m15/m30,nor above h4/d1.
Since you can`t post a picture of spectrum analyzer,try to post the next valleys and next peaks too,of the cycles you find.
Gbpusd Analysis
Look at the 2 pics..one is the spectrum analyzer for gbpusd h4,the other for gbpusd h1..full history in my pc..around 3000 bars for each..there is a clear peak at 14 period in h4..and another at 56(4*14=56) periods in h1.
Now we know what to look for
..next post will follow with actual implementation.
Implementation
I created 1 cycle based on 56 peak at h1,and 2 cycles,based on 14 periods peak at h4..all for gbpusd.
Then I put them into a gbpusd chart,and used a mtf indicator graciously made by Perky and posted at the SimbaConMan thread to use h4 on h1..colours red and blue..and then ,without mtf(obviously)I attached the h1 cycle..colour gold.
Looks interesting..doesn`t it ?
Stlm2
And,if we add the STLM2,it looks even more interesting..looks like today we could look for a good sell spot...of course in demo,do not try it at home
.
Since this is probably something new to most of you,and having being a newbie myself,I can imagine some of you rushing to short the cable and phoning your friendly Ferrari dealership to ask for colour assortments and availability dates...DO NOT DO IT..you don`t know this technique,you don`t know its potential pitfalls..and,until you get more experience with this,on demo,you can not decide if this is worth including in your trading toolskit or not...
What I ,personally do with this info is first..checking the inmediately higher timeframe..h4..with a STLM2,which,lately has become my best friend in all matters trend related,specially in h4,since this tf ,usually, gives trends that last approx a trading week 4 to 6 days..and,surprise,surprise,looks like we are in a potential retracement during the beginning of an uptrend..so,the safest way to use these cycles,would be to wait for the completion of the retracement aka "when the blue and red turn up"in h4..and then look,with your usual trading strategies,for a good buy spot...
My personal conclusion is:
1-weekly trend is up and still young..as defined by stlm2 h4
2-We are in a potential retracement that will last,probably and judging by the average length of the cycles ,until Monday
3-There seems to be a good short opportunity,countertrend,in H1,NOT YET CONFIRMED BY PRICE ACTION..so high risk
Now I have 2 options in my trading menu:1-Either I trade both setups..or 2-I wait for Monday,recheck,and if everything still points to an uptrend and cycles look ok,I go long with my usual methods,looking for a buy spot at m5,m15.
So,these cycles and digital filters are not the magic Holy Grail,but they are interesting tools that can give us an edge.
Lol....i know understand what i'm doing....i need to use the whole history :-)....no wonder i was scratching my head going 202-210 bars....
So, to clarify....SATL, RSTL, FATL, FSTL (which lead to stlm ftlm) are "shorter analysis time frames" (200-220 bars or so) where as cycle analysis is longer time frame, hence full histories...
makes perfect sense now...off to work, but i shall be back :-)
Thanks for sharing Simba. I think i speak for everyone when i say it's people like you that help people like me continue to move forward, create, question and learn.
cl
I'm hoping to post more later....but here is a pic of the 1 hour cycle....Pretty good :-)...
The only thing i'm noticing is that on my 4hr analysis of GBPUSD, i only have about 1800 bars of data....not 3000....the 14 does not come through at all...it's more like 12....i'm thinking by downloading some historical data and doing the needed conversion it will be easy to get it there...
More later,
cl
longer timeframe
Lol....i know understand what i'm doing....i need to use the whole history :-)....no wonder i was scratching my head going 202-210 bars....
So, to clarify....SATL, RSTL, FATL, FSTL (which lead to stlm ftlm) are "shorter analysis time frames" (200-220 bars or so) where as cycle analysis is longer time frame, hence full histories...
makes perfect sense now...off to work, but i shall be back :-)
Thanks for sharing Simba. I think i speak for everyone when i say it's people like you that help people like me continue to move forward, create, question and learn.
clYou can use both strategies for both sets of indicators,there are no limitations,but to use what works and discard the rest.
In the case of trying to find a persistent cycle among timeframes,it makes more sense,IMHO,to look for time persistence too,so,that is why I use full history.
BTW the sell setup didn`t work on Friday even if still has a few hours to go during next week`s opening,and the real interesting thing would be to see
its turn to long on Monday in sync with the 4h trend.
This is a pic of the EURJPY H1 cycle,done with the second method,that I started to use Live on GBPJPY M5 yesterday(after testing,of course
Yesterday I closed 3 trades with,among other tools,this cycle..and today I just closed the first trade of the day,I want to use this cycle as a tool in low range days to look for 20 to 50 pips per trade with few lots than usual,Trade length from 20 minutes to a few hours,just to "cover the expenses and pay my salary"..
The trick is to look at other confirming tools,like support area approaching when downcycle is "mature",etc,it is an art,not a science,but this tools help a lot..
Details of my trades,I have edited the last 4 digits of the order number,and times at my broker(GFT) are always GMT,Profits,both trade by trade and total,are in Euros...Additionally,FFL,that was trading the same setup,since I am mentoring him,can confirm and post his trades if he wishes to do so
,he had some issue with internet connection today,so,today`s trade will be probably different,but yesterday ones are practically the same,with small differences since he uses another broker.
Feb 20, 2008 7:50:40 AM GBP/JPY DDL B 300,000.00 CR 209.53 62,859,000.00 DB Feb 22, 2008 8:00:00 PM 20690XXXX
Feb 20, 2008 10:24:13 AM GBP/JPY DDL S 300,000.00 DB 209.91 62,973,000.00 CR Feb 22, 2008 8:00:00 PM 20691XXXX715.92 CR
Feb 20, 2008 10:57:28 AM GBP/JPY DDL B 300,000.00 CR 209.89 62,967,000.00 DB Feb 22, 2008 8:00:00 PM 20691XXXX
Feb 20, 2008 5:27:08 PM GBP/JPY DDL S 300,000.00 DB 210.34 63,102,000.00 CR Feb 22, 2008 8:00:00 PM 20695XXXX 847.80 CR
Feb 20, 2008 7:05:04 PM GBP/JPY DDL B 300,000.00 CR 209.94 62,982,000.00 DB Feb 22, 2008 8:00:00 PM 20696XXXX
Feb 20, 2008 7:31:36 PM GBP/JPY DDL S 300,000.00 DB 210.29 63,087,000.00 CR Feb 22, 2008 8:00:00 PM 20696XXXX659.40 CR
Feb 21, 2008 7:31:33 AM GBP/JPY DDL S 300,000.00 DB 210.36 63,108,000.00 CR Feb 25, 2008 8:00:00 PM 20707XXXX
Feb 21, 2008 8:44:51 AM GBP/JPY DDL B 300,000.00 CR 210.16 63,048,000.00 DB Feb 25, 2008 8:00:00 PM 20708XXXX 376.80 CR
Net 0.00 CR GBP 1.3208
JPY 0.00628
2,599.92 CRPlease excuse the tardy reply.
I meant to address this post sooner but was experiencing some technical difficulties that Simba can and has verified. 
Then the posts moved past this page and it slipped my mind.....I digress.
I can, indeed, attest to the validity of these trades. Every one is 100% accurate. I was unable to place every trade for fear of aforementioned continous and prolonged disconnecton issues, but Simba was sending IM's real-time so that I could have proof of the order
Thanks for sharing Simba. I think i speak for everyone when i say it's people like you that help people like me continue to move forward, create, question and learn.
clYou are, indeed, preachin to the choir. Simba has taken me under his wing, or paw I should say, for no reason other than to express his true benevolence. As a result, my understanding of the markets as well as myself have drastically improved. I still have quite a journey ahead of me, but w/ a guide like Simba, I'll be able to shave off hundreds of "pointless miles" and save thousands in "wasted gas". I feel deeply indebted to him for all of his teachings and friendship. I will figure out a way to return the favor.
Peace,
F.F.L.
P.S.
If I keep at it, I will one day be in a position to use the same broker. (DukasCopy).
FFL,
Definetly keep trying. In my opinion you should be optimizing your digital filters so that you get few coefficients. If you have an SATL that has 40 coefficients, it's going to be slow in terms of processing. You then have to delay it by 20, which will in effect make a useless STLM. What i have found through testing is that the same SATL that gave you 40, if you tweak p1 or d1, you can get it to less than 20 by putting in different peaks and have it "look" very close to the SATL with 40 coefficients. It's basically guess and check. I'm trying the mtf spectrum that i talked about earlier so we'll see how that goes. What i was thinking about doing was this - Get the data from the first week of january, then run the EA with digital filters for the second week of Jan. Record results. Then get the data for the second week, make filters, reoptimize the EA, and run it against the 3 rd week...etc etc. I'm finally getting some free time again, so hopefully i can proceed with that. We shall see.
clYo Clanh04,
Thanks for the tip on the coefficients. This alone improved my output of DF by leaps and bounds. I still have much to learn. You and Simba are talking cirlcles around me. LOL. From what I've seen, there isn't much documentation out there about using digital filters in trading and it seems to be a little known about topic. So the fact that there are individuals like you guys in one forum presently posting ont the topic with whom one can bounce ideas off and improve one's understanding is quite a stroke of luck, to say the least. One of the few good threads left on TSD.
Any other readings anyone would recommend on using digital filters in financial markets?