Discussion of article "Application of the Eigen-Coordinates Method to Structural Analysis of Nonextensive Statistical Distributions" - page 3
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
Interesting material, thank you. I don't want to disturb the reigning mathematical nicety here, but I still can't help asking two simple questions:
1. The question of the practical value of these distributions. What should we come to as a result? Description for its own sake is fine, but (I apologise, of course) it smells of botany.
2. Is it reasonable to try to describe completely different in nature processes occurring at different "levels" on the market by a single distribution. The problem of "kinks" has already been mentioned here, but this is only a part of the problems that exist. Moreover, the composition of the processes itself changes significantly in different historical time intervals, how you want to describe it by a single distribution - I don't understand.Thank you for your interesting questions, I think they are far from simple.
Mathematical goodness is not our goal, the hope is that statistical properties can provide information about the market. Each particular instrument has its own properties, if they turn out to be constant then that is its characteristic. It was after the discovery of invariants that major breakthroughs in understanding the world came about.
1. The distribution shows on which side the statistical advantage lies. For example, with the SP500, the distribution is shifted to the right and the graph shows a trend.2. The only criterion of truth is experiment. But it can be used not only to confirm the reasonableness of theories, but on the contrary - we need to make sure it gives us a kick in the hat - it is the only way to go beyond existing ideas - it is important to find where exactly we are wrong. In this sense distributions can tell us where they lose their meaning - you have to count and look.
I don't know the right answer, I don't have the crystal ball (looked everywhere), one of you must have it.
"Historians of such ancient intellectual disciplines as poetry, music, painting, and science praise the role of eminent practitioners whose achievements have enriched the experience and broadened the perceptions of the devotees of these disciplines, and awakened and strengthened the talents of the followers. What new things they have contributed are based on a combination of virtuoso practical skill and insightful comprehension of fundamental principles..."
C.A.R. Hoare (from the foreword to the book The Discipline of Programming)
Let's share experiences, test ideas and discuss results - together it's much easier to find the right answer.
Thanks for the interesting questions, I think they are far from simple.
Mathematical goodness is not our goal, the hope is that statistical properties can provide information about the market. Each particular instrument has its own properties, if they turn out to be constant, that is its characteristic. It was after the discovery of invariants that major breakthroughs in understanding the world came about.
1. The distribution shows on which side the statistical advantage lies. For example, with the SP500, the distribution is shifted to the right and the graph shows a trend.2. The only criterion of truth is experiment. But it can be used not only to confirm the reasonableness of theories, but on the contrary - we need to make sure it gives us a kick in the hat - it is the only way to go beyond existing ideas - it is important to find where exactly we are wrong. In this sense distributions can tell us where they lose their meaning - you have to count and look.
I don't know the right answer, I don't have a crystal ball (looked everywhere), one of you must have one.
Let's share experiences, test ideas and discuss results - together it's much easier to find the right answer.
Yes, it is of course the trump jack, but I would like a queen and a king.
How can the distribution answer the following questions:
Having in our arsenal methods that answer these questions, we can consider that the transition from academic research to practice is complete.
Although I have vague doubts that the distribution also answers the question of trend definition.
In another thread I gave screens on which you can clearly see an emerging trend within an older trend, those small movements that fit into the old trend and do not lead to statistical perturbations, subsequently become the first reference line to describe the new trend.
I think that trend nesting is a very difficult problem to recognise and worth doing.
Because recognition methods can greatly move the process of understanding the evolution of a trend and give an answer to the question that everyone implies, but no one has yet given the answer " what is a trend?".
It is a paradoxical situation, everybody sees it, but nobody knows its definition, it is like a UFO :)
In another thread I have given screens that clearly show an emerging trend within an older trend, those small movements that fit into the old trend and do not lead to statistical perturbations, subsequently become the first reference line to describe the new trend.
Quantum:
Each particular instrument has its own properties, and if they turn out to be constant, this is its characteristic. Serious breakthroughs in understanding the world appeared exactly after the discovery of invariants.
Alas, but the properties of instruments are constantly changing. There are very few properties, any patterns that last for years.
Quantum:
1. The distribution shows which side has the statistical advantage. For example, with SP500 the distribution is shifted to the right and the graph shows a trend.If you take a time period where the SP500 first goes down and then up, you will probably (it is a good idea to check) get a symmetrical curve.
There is no trend here, if you look at the distribution, and meanwhile there will be two great opportunities to make money, first shorting and then longing. This is the disadvantage of distributions, they have no concept of time, or development of the process in time. And in the market it is very important to watch the dynamics.
Of course, it is possible not to take the whole instrument, but to cut it into intervals and build distributions for them, which will result in something similar to a set of market profiles (something like a candlestick chart, but instead of candlesticks there will be distributions). In this case, we will again come to the well-known problem of process breakdown. More precisely, the task of determining the process fault as soon as possible.
Quantum:
2. The only criterion of truth is experiment. But it can be used not only to confirm the reasonableness of theories, but on the contrary - we need to make it give us a kick in the hat - it is the only way to go beyond the existing ideas - it is important to find the place where exactly we are mistaken. In this sense, distributions can tell you where they lose their meaning - you have to count and look.
Where, in which thread, are such drawings posted?
For some reason the principle of falsifiability comes to mind.
Look not for what confirms, but for what disproves.
Haven't read the article yet.
The attachment file can't be download. Please check, the link show error like this :
403 - запрещено. Доступ запрещен.
Предоставленные учетные данные не дают права на просмотр этого каталога или страницы.
The attachment file can't be download. Please check, the link show error like this :
Fixed. Thank you.
MetaQuotes,
Can you translate the discusions of the article in Russian to English, because there are some practical applications.Google translator is no good.