What is everyone looking for? - page 18

 
faa1947 писал(а) >>

Well, brother bent. Volatility is the difference between price and average. It is the higher frequency component of BP. the vlet decomposes into these frequencies, the first of which is the trend. The sum of all this is the initial BP. There is an orthogonality proof for this.


Volatility is the price change over a fixed time interval. And that's it - without any trend, waving or subtraction. But the swing change will correlate significantly with it, especially if the swing is with period=1 :)
P.s. The correct spelling is "volatility"
 
Avals писал(а) >>


Nope, volatility is the change in price over a fixed period of time. And that's it - without any trending, waving or subtraction. But the change in swing will correlate with it significantly, especially if the swing has period=1 :)
P.s. The correct spelling is "volatility".

In ATR it is, in BB and channels it is not. The standard measure is the RMS. But this is an empty conversation. The whole topic is flubber. All attempts to push the topic in any direction, with supporters, evidence, literature have failed. Adieu.

 
faa1947 >>:

Все попытки столкнуть топик на какое-либо направление, имеющую сторонников, доказательства, литературу ничем не увенчались. Адью.

Hang on, take your time. The topic is turning into something interesting, not just a splash of perplexity of the topic-starter. And you have contributed to it too.

faa1947 >>:

You cannot analytically derive another indicator from one indicator

.

You cannot get volume indicators from a trend indicator and you cannot get volatility indicators from both of them. Hotf all can be obtained from the original BP.

Oh how interesting. Anyway, my understanding is that orthogonality of indictors is in some way just functional independence. Right, Alex?

And in general, tell us more about the culture. Or give some links.

 
Mathemat писал(а) >>

Hang on, take your time. The topic isn't just turning into an outburst of perplexity by the topic-starter, but into something interesting. And you have contributed to it too.

Oh, that's interesting. Anyway, my understanding is that the orthogonality of the indictors is, in a way, just functional independence. Right, Alex?

And in general, tell us more about the culture. Or give me some links.


Um... M, I offered a clear criterion for assessing the quality of an indicator in the first post. All the rest is just nonsense. I am not interested in anything except methodology and its estimation and preferably its application to display my own indicator developments. I look through this indicator and then I see that it is something real comparing it with other indicators. The rest is just chatter about nothing.

 
Mathemat писал(а) >>

Oh, that's interesting. Anyway, my understanding is that the orthogonality of the indictors is, in a way, just functional independence. Right, Alex?

And in general, tell us more about the culture. Or give some links.


Oh, and this one too, forgot to add - sorry.

 
SProgrammer >>:


Хм... М, я в перовм сообщении предложил четкий критерий оценки качества индикатора. Всё остальное просто мутотень. недоумение спровоцировало блуждание в потьмах половины форума. Мне вот ничего кроме методики и ее оценки и желательно применения при выкладывании собственных разработок индикаторов неинтересно. Сделал челоек индикатор - получил по данной методике результат - приложил к индикатору - я просматривая этот индюк - глянул - ну типа ага что-то стоящее, В СРАВНЕНИИ с теми что есть другие. Остальное просто болтология ни о чем.

No, it's fine, S. I'm not judging your style.

But your criterion you... how shall I put it gently... you didn't get it right. If you're interested, I can be more specific about what I meant. It's not easy to create such a criterion.

P.S. Thanks for the codes, by the way.

 
Mathemat писал(а) >>

No, it's okay, S. I'm not judging your style.

But your criterion is... how shall I put this delicately... you didn't get it right. If you're interested, I can be more specific about what I meant. It's not easy to create such a criterion.


Well did you look, didn't I even give the code there? -So that I could, importantly, without looking at.... I mean having only zero bar, to calculate some indicator, on the basis of some recommended by the author of this indicator - the methodology of using this indicator - to compare it with IDEAL .... How to compare - you're probably better at it.

It would be very interesting of course. :)
 
Let's make a long story short. Let's say we have a particular system - two wands, entry/exit by crossovers, no filters. We test and analyze the history of deals.
1. Trade 2367: Entry. Taps crossed in the way that the system requires to enter long. Entry. The analysis of price behaviour shows that it went down by 17 points after entering. How to assess the quality of entry?
The price could have behaved differently: a couple of minutes after entry it went up by 11 points. And what is the quality of entry?
2. Trade 2367: Exit. The wagons have crossed again. We interpret this as an exit from the long position. Exit. The maximum paper profit in the trade was 38 pips, but they crossed already at a paper profit of 22 pips. How do we evaluate the quality of the exit?
3. How do we evaluate the quality of this trade as a whole?
4. How to evaluate the quality of the system as a whole?

P.S. What can an ideal have to do with this system? The ideal is subjective and probably built on a different idea...
 
Mathemat писал(а) >>
Let's make it short. Let's say we have a particular system - two wheels, entry/exit by crossovers, no filters. We test and analyse the history of trades.
1. Trade 2367: Entry. The bars crossed in such a way that the system requires to enter long. Entry. The analysis of price behaviour shows that it went down by 17 points after entering. How to assess the quality of entry?
The price could have behaved differently: a couple of minutes after entering it went up by 11 points. And what is the quality of entry?
2. Trade #2367: Exit. The wagons have crossed again. We interpret this as an exit from the long position. Exit. The maximum paper profit in the trade was 38 pips, but they crossed already at a paper profit of 22 pips. How do we evaluate the quality of the exit?
3. How do we evaluate the quality of this trade as a whole?
4. How to evaluate the quality of the system as a whole?

P.S. What can an ideal have to do with this system? The ideal is subjective and probably built on a different idea...


That's not what I'd do - I'd use some kind of "manufacturer's recommendation" ( and there are definitely some for mashes, verse and folk music) to get time-distributed signals, and compare them to IDEAL's signals. As far as they diverge (0 to 1, in % is possible) the tested IDEAL will fail. That is you must first obtain the signals.

What tools to select - here it is up to the "author", he should probably select the parameters of the zigzag and his ( being tested) indicator.

** By the way, I've corrected the code - there was an error. :)

 
Mathemat писал(а) >>
P.S. What can an ideal have to do with this system? The ideal is subjective and probably built on a different idea...

But how do you evaluate "coincidence"? dunno.