TA - töötab siis või mitte?
Kommentaari jätmiseks loo konto või logi sisse
-
See küsimus on käinud lhv foorumitestki mitmeid kordi läbi.
Isiklikult ei oska ma midagi arvata moment.
Juhul kui TA peaks üldjoontes toimima intraday, siis oleks ainuke selle mõtekas rakendusala futuurid või forex.
Võib olla on mõned erinevused aktsiate ja futuuride TA-l ja võib olla töötavad mõned meetodid ainult aktsiatega.
Jätame välja TA mida tehakse uudiste perioodil.
Asi mis huvitab, on see, kas keegi on võimeline TAga järjepidevalt teenima.
LHV foorumitest pole vähemalt silma jäänud, et Eestis leiduks mõni TA täisproff scalper ja kas neid üldse on.
[abesiki] i am thinking - is it possible to make profits with TA
[abesiki] it seems quite impossible
[sean] i do but it took a while to learn that less is more
[Musashi] time better spent moving in the direction of the Profile
[abesiki] how long have y been profitable with your TA methods?
[sean] about 2 years
[abesiki] that`s quite good. intraday scalping , newsplay or smth else?
[sean] the learning curve is a hard one
[timing] whats t a method sean?[sean] i do intraday ,swing and intermediate term trading
[sean] i only use 3 indicators for intraday tarding and i use 4 set-ups to trade
[abesiki] could y say how big is your procentual return on account per year?
[sean] i use 2 exp. MA for trend , bollinger bands for range and stochastic of RSI for an oscillator
[indyyoga] what period ma's?
[sean] 4,14 EMA on 13 min chart ,3 min chart for entry /exit ,60 min chart for bias
[sean] Bollinger band set at 13 ,2.618
[sean] i trade the 13 min chart because i lose money trading the short time frames
[indyyoga] How'd you come up with 13 min as opposed to 10 or 15?
[sean] something I read in a book by Dr. Elder and the difference of 5 betwen time frames
[ThurstonHowell_3] sean I used to trade with the 13 min chart too, but somehow I switched over to the 15 min
[sean] 3,13,60,day,weekly
[enthios] sean, you sound like an acolyte of RAL - do you know him by any chance?
[sean] no
[enthios] ok
[enthios] he was my first Fib teacher and used similar setups
[sean] i just try to keep my trading simple and focused
[enthios] 13min
[enthios] bollingers at 13, 2.618
[enthios] ("fib" settings)
[indyyoga] was gonna say 13 and 2.618 sound like fib's
[enthios] your trading setups sound excellent - simple - and a good combination of indicators as well
[sean] it shows the extremes of range at those settings
[enthios] trend + oscillator + range
[enthios] I use one more as well... just look at the market profile for clear pivot points
[lolly] 390 trading min 13 evenly divisible
[indyyoga] 390 minutes in trading day?
[enthios] lolly, although the futures has 405 mins
[lolly] 9:30 - 4:30
[enthios] so 135-min chart is useful in that regard
[lolly] 4:00
[enthios] divides trading day into 3 equal periods
[indyyoga] you 're all mathemeticians in here... I'm just lookin for a dt or h & s!
[enthios] i don't use 13 min any longer...but i do incorporate longer term indicators into my short term chart, so i only need to look at one chart. i trade off a 360-tick chart (similar to 1~3 mins) but because the indicators come from higher time frames, I only get 1~2 trades per day
[enthios] abesiki, just read your earlier post...yes it is possible :)
[enthios] For example, taking sean's two ma's of 4 and 14 on a 13-min chart for trade entry and on 60-min chart for bias, you can put those four ma's on a 13-min chart. The 4,14 on 60-min chart simply converts to a 18,65 on the 13-min chart. Easier than looking at two charts, and uses less screen real estate.
[ThurstonHowell_3] hod tick 1327
[sean] concentrate on one set-up ,say MA cross in the direction of the larger trend, and practice it until you "own"it
[enthios] TH3, we have a virgin poc (vpc) at R2
[ThurstonHowell_3] ty
[enthios] also a natural target :)
[indyyoga] i was doing fib's on the 1 min. On daily, I got 10327 as .786 of last down move.
[abesiki] thanks guys
[huskergal] i do that too enthios
[sean] hope it helps -
tehniline analüüs ei ole midagi muud, kui statistika. sinu küsimus on kohatu. võiks ju küsida "kas kindlustusega on võimalik raha teha?"
statistika enamasti toimib, lihtsalt hälveteks peab valmis olema. -
aitäh muidugi algtõe meeldetuletamise eest, et tegu on statistikaga.
backtesting ei projekteeri sulle reaalseid tulemusi ette, mida suuda saavutada.
testi näiteks suva indikaatori erinevaid variante viimase kuue kuu lõikes, mis annab sulle näiteks X punkti. Kas see variant on sul sama tootlik ka järgmised 6 kuud reaalis? Enamasti mitte.
Kübe konarlik on järeldada, et kuna viimase viie aasta jooksul on kindlustused kasumit teeninud, siis järelikult teeniks järgmised viis ka uus turuletulija. -
Backtesting lõhnab ikkagi statistika järele. 6 kuu kaupa pole erilist mõtet backtestida, tuleks backtestida ikka näiteks 20.a. ja võimalikult palju erinevaid turuolukordi sisse sinna mahutada. Näiteks 2003.a. backtestimine ja 2004.a. backtestimine annaks ilmselt väga erinevaid tulemusi.
-
Tõesti bactestinguks vaja pikemat horisonti. See mis võis töötada 2003 ja varem nagu õlitatud kell, ei tee seda mitte 2004. Tuleb olla paindlik oma systeemidega.
Yks väike arvamus pikaajaliselt turgudel tegutsejalt.
A great deal of S&P500 trading is surviving. Beginning traders can only think about Profits-Profits-Profits... just like a gung ho new Cop on the beat or a soldier. Gotta get those criminals or be like Rambo.
But you have to realize, like a soldier in a danger zone, or a cop working the Ghetto, that it isn't always about just this big fantasy of being the big Bucks trader but in many, many cases in the real trading world...it's about SURVIVAL. Being able to trade again when things become more fluid, more obvious, when the setups pan out better rather than just flipping around in 3 point ranges or stalling for hours on end in a 2 point range. That is what we are being handed in the recent market environment every day now. Many, many established traders have been wiped out in 2004 and we have survived the massive market changes. -
Capital preservation e. kapitali säilitamine peaks olema tõesti iga kaupleja esimene reegel.
-
Mainisingi, et backtesting on statistika.
Aastate horisonti vaatavat backtesti tehakse klassikalistele turuolukordadele.
See kõik on nii swingtreidide puhul, et võetakse väga palju erinevaid turuolukordi.
Tehakse klassifikatsioonid jne..
Niisuguse probleemiasetuse korral pole mittemingisugust probleemi, et kas TA töötab või mitte - loomulikult peaks funktsioneerima kui hea idee leiad.
Küsimus on selles, et kas TA töötab nii öelda puhtal kujul - skalpimisel.
Nii palju kui mina olen aru saanud, siis skalpimisel pole enamasti sul mitte munagi teha aastataguse statistikaga, 10 aasta taguse võid kohe prügikasti saata.
Isegi viimase kuu andmed võetakse üksipulgi lahti ja kohendatakse.
Üks ainuke päev kahjumit ja kohe mõeldakse, et midagi on mäda.
Kui 50 tehingut keskmiselt päevas ja 20 päeva, siis 1000 tehingu backtest juba näitab toimimise kohta midagi.
Aastaga muutub skalpimisel palju nagu räägitakse.
Kõike ei saagi tavasoftiga testida - nii mõnedki on lasknud või on ise endale progenud C++s või C# spetsiaalsed programmid.
Tradestationist järgmine level peaks olema www.smartquant.com -
Ehk väikseks abiks järgnev http://www.deltat1.com/Examples/Aug2004/Aug2004-5min.htm
Kuidas liikus ESU4 RTH ajal. Kui veidigi syveneda, võib välja lugeda väga sarnaseid olukordi igal kauplemispäeval. Oleneb ainult sellest, kes mingi nurga alt vaatab ja kui keeruliseks oma kauplemise ajab. -
Kindlasti saab TA-d ka scalpimise puhul kasutada - lost art of tape reading.
-
Tehnilise analüüsi töötamise või mittetöötamise kohta saame vastuse sellel sügisel. Gary B. Smith on genereerinud ühe huvitava graafiku :-)
-
Currani arvamus
In the everyday world, we take our specific problems to a specialist. You want a mechanic who specializes in the particular make and model to work on your car. You seek the opinion of an orthopedist for a problem with your knee. To address specialized problems, we look for experts with specialized knowledge, information, and training. The same goes for trading. In trading however, the trick lies not in finding an expert to address your problem, but in becoming that expert.
Once you've decided to trade, you must determine the kind of trader you want to be. What kind of time and energy do you want to commit to trading? What are your personal reward goals? Do you think that you should be a day trader? You may find that it's too stressful during the day if you have to sit in front of a monitor and make trades every 15 minutes, when you could be doing other things you enjoy. In other words, your trading strategy must fit with your personality. Many traders have a broad understanding of many approaches to trading, but most rely primarily upon technical analysis as the basis for their short-term decisions. However, technical analysis is a vast body of research and literature, covering everything from simple chart patterns to complex computer-based neural network systems.
Personal interests may play a roll as you narrow the field of technical-based approaches to the one most suited to you. For example, if you have an engineering or mathematical background, a statistical system such as Moving Averages and Bollinger Bands will likely make more sense to you. If you are a visual person, then recognizing patterns such as triangles, wedges, and Butterflies may come more easily.
There are 2 schools of thought: The first says that it's best to focus on a group of securities, such as a basket of futures contracts or stocks, and learn how they move and what forces drive their movement. The second school says "trading is trading" and everything can be traded the same way.
