Classical thechanalysis doesn't work any more. What works, maybe quantum? - page 16
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
What would be the variance of a non-stationary series? Is it even possible to calculate it? Same thing about returns.
It depends on the length of the row. You can, of course, calculate it. I have already calculated it - for an infinite series the variance equals infinity. The same is true for the yield: The yield of a portfolio containing all marketable assets on an infinite investment horizon equals plus infinity. Is there any doubt about this?
It depends on the length of the row. You can, of course, calculate it. I have already calculated it - for an infinite series the variance equals infinity. The same is true for the yield: The yield of a portfolio containing all marketable assets on an infinite investment horizon equals plus infinity. Is there any doubt about this?
There is no need to count for infinity. It is enough to accept the fact that for a non-stationary series the characteristics will be unreliable. This is enough to conclude that there is nothing to substitute in reality for Markowitz's remarkable valid formulas. After that, either shove the Nobel laureate far back on the shelf, until better times. Or try to modernise the theory.
You don't have to count for infinity. It is enough to accept the fact that for a non-stationary series the characteristics will be unreliable. That is enough to come to the conclusion that there is nothing to substitute in reality for Markowitz's remarkable valid formulas. After that, either shove the Nobel laureate far back on the shelf, until better times. Or try to modernise the theory.
For any series the characteristics will be unreliable, the only question is the degree of unreliability. Markowitz's strength is not in his formulas, which simply cannot be true due to his simplistic approach, but in the ideas behind them. One of the ideas is that TA is impossible.
I accidentally came across this thread, I only read 1 page, I didn't have time for more. Question to everyone who thinks TA does not work, do you not use any indicators? If you do, don't lie, the indicators work based on TA. And there is no need to fool yourself and others!
Didn't immediately notice the thread, and the time was extremely short, so I'm trying to catch up :o). I completely agree with the author, technical analysis has not just stopped working - it has never worked for the future at all. Under TA I understand absolutely everything that can be classified this way, starting from the "TA" section of this site and finishing with all kinds of bullshit, like "head and shoulders", arcs, triangles, Fibo (this is a separate bullshit) indicators, levels ... etc. etc. I'm sorry for the time I spent studying this rubbish, but apparently it was necessary, it's evolution however. Studied thoroughly, for a long time, each time trying to understand the "nature" of the inoperability of each of these methods. The particular nonsense is waves, in all their manifestations.
All you have to do when working with any TA method is to stop and try to answer the questions "what am I doing", "why am I doing it", and "how is it working?". The first one is easy to answer, but the second and third are a bit more difficult if you describe the result rather than the drawing.
For any series the characteristics will be unreliable, the only question is the degree of unreliability. Markowitz's strength is not in his formulas, which simply cannot be true due to his simplistic approach, but in the ideas behind them. One of the ideas is that TA is impossible.
I haven't read such an idea in his book. And I don't see how it relates to his theory of the efficient portfolio. Can you give me a link?
I accidentally came across this thread, I only read 1 page, I didn't have time for more. Question to everyone who thinks TA does not work, do you not use any indicators? If you do, don't lie, the indicators work based on TA. No need to fool yourself and others.
Price is also an indicator. The question for you as a TA expert is: "What are the basic principles on which TA is based?"
It is not necessary to count for infinity. It is enough to accept the fact that for a non-stationary series the characteristics will be unreliable. It is enough to come to the conclusion that there is nothing to substitute in reality in Markowitz's remarkable valid formulas. After that, either shove the Nobel laureate far back on the shelf, until better times. Or try to modernise the theory.
Unsteady random time series can be described by a formula, but that does not mean that they can be predicted. To say that there is nothing to substitute in reality is utter nonsense.
Price is also an indicator. The question for you as a TA expert is: "What are the basic principles of TA?"
I assume that the most important and most deceptive one is the belief that if something is drawn by the indicator, then this something will take place for some time (so that the picture of the indicator at the current moment gives hope for similar behaviour in the future). This "something" could be a channel, a slope of a muving, RSI value close to 80, etc.
In principle this hope is justified in the case of, say, a moving average with a long period: the derivative of a moving average will not just change its value abruptly. It will change smoothly. But it won't help, because for such a moving average the contribution of the current price is too small. It may change by hundreds of points (old), and the muvinng will not move at all.
Well, of course, there are a few other postulates. Like this one: "price reflects everything".
P.S. In other words, if we want to make correct smooth indicators, the contribution of the current price should be as large as possible. So how do we do that?
I suppose that the most important and the most deceptive - is the belief that if something has been drawn by the indicator, then this something will take place for some time (so the picture of the indicator at the current moment gives hope for the same behavior in the future). This "something" could be a channel, a slope of a muving, RSI value close to 80, etc.
In principle this hope is justified in case of, say, a wide period muving: the derivative of a muving can't just change its value abruptly. It will change smoothly. But it won't help, because for such a moving average the contribution of the current price is too small. It could change by hundreds of pips (old) and the muwwing would barely move.
Well, of course, there are a few other postulates. Let's say this one: "price reflects everything".
Naturally not. Probability at any point in time 50/50. Ie there is no inertia, but not all so categorical, you can temporarily detect with stat. significance weakly effective markets, but with them as a goat milk.
If you take the ACF of the muvinga,it will be positive, that is, as if we can predict the shift m.o. series, but it's self-deception. In general, all the technical analysts caught on the effect of false regression - someone immediately earn a bunch of dough, and then also plummet, someone also plummet.
"Price is everything" - in order to understand the validity of this postulate, you have to answer the question, "What is everything?"