Combining multiple indicators in an EA - page 6
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, Peter came along and ruined everything. Actually, he said it right.
He didn't tarnish anything... but someone will use the search, someone will find it, someone will spit out the husk...
There he is, Svinozavr in all his glory, not hung over. He's going to give you all a good kicking, you young Stalinist Michurins. You'll know how to put the dice together properly...
Клуб юных мичуринцев, блин. Лысенки недоученные. ))) Вы какую "развесистую пшеницу" объединяя/скрещивая хотите в результате получить? Экспертостроение при дворе императора Рудольфа. Фаусты доморощенные. // все, успокоился...)))
Индикаторы - это аналитический инструмент ТА. Если вы, не соображая, ЧЕГО они меряют, начнете их бездумно комбинировать, то и рез-т будет соответствующий. Для начала нужна идея, какое состояние рынка, какие условия вам хочется вычленить, а дальше уже идет подбор аналитических инструментов - индикаторов.
А так... "я это добавил, то-то выкинул... стало лучше... хуже..." С бубном канкан не пробовали?
You should not take it out on us nerds. The topicstarter here teaches.
By the way.
No one has refuted the stage development theory yet.
And the problems of potato degeneration in the southern regions remain.
Needless to say - forecourt education is as bad as vetting...
Just got the essence of it, the epiphany dawned! - BOOM! Kolyan is here.
Or Svinozavr ;)
If you don't have a clue WHAT they're measuring, start mindlessly combining them.....
Why do you think so badly of people, Svinozavr.
AlGor,how do you propose to apply it, do you have any ideas about it?
ничего не опошлил... а ведь кто-то воспользуется поиском, кто-то найдет, кому-то шелуху придется сплевывать...
Weren't you Simon the other day saying:
can't wipe everyone's snot off
...♪ the rose-colored glasses won't come off ♪ no one can be saved from the bumps... Only one's own mistakes teach one anything, although forex takes advantage of the fact that a trader who made a mistake once will repeat it over and over again...
Здесь топикстартер учит.
The topikstarter here is not teaching, but misleading:) as the conclusions he draws and even more so the calculations are generally incorrect. For example, the conclusion
3. the incorrect connection of several good indicators will not improve the system;
is correct only in the case when readings of "good" indicators often contradict each other (and it can be shown strictly mathematically). And the probability of the correct forecast with "connected" indicators can be estimated only on the basis of historical data. The formula, if necessary, I will write. You could even call it the "utility formula":))))
... correct only if the 'good' indicators often contradict each other...
>> Exactly.
AlGor,а как вы предлагаете его применять, у вас есть идеи по этому поводу?
I don't. However:
...But according to our compatriot, physicist Sergei Maslov of Brookhaven National Laboratory, if you allocate capital between two unprofitable, losing stock portfolios, it will increase, not decrease...
Since I promised, here's the utility formula, let's have it. (Actually, it's usually called Bayes formula:))
P(A+ AND B+|M+)*P(M+)
P( M+|A+AND B+)= --------------------------
P(A+ AND B+)
P(A- AND B-|M-)*P(M-)
P(M-|A- AND B-)= --------------------------
P(A- AND B-)
Notation of events: M+ (M-) - market went up (down), A+, B+ (A-, B-) - the corresponding indicator gives the "up" ("down") signal
The vertical line means "subject to".
The first probability is a probability of profit at consensus prediction (prediction according to I system, as topisstarter called it) for buying, the second one is the same for selling. In a symmetrical market P(M+)=P(M-)=0.5 The remaining probabilities can be estimated by the frequencies of corresponding events, counting them from the history, for example, P(A+ AND B+|M+) can be replaced by the frequency of the event of joint Buy Indicators A and B given that the forecast was correct, P(A+ AND B+) - is how often the indicators show the same (in this case "up")
Note that if indicators are built symmetrically, then both formulas must give almost the same value, it will be an estimation of the total probability of profit