How to form the input values for the NS correctly. - page 4
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
By the way, here's a good rummage through the forum, found some interesting posts suggesting inputs
https://forum.mql4.com/ru/8835/page2 by plan
и
https://forum.mql4.com/ru/9321/page18#51761
2 Sart - if you are a beginner, you might be interested in the code from my post from https://forum.mql4.com/ru/12474/page9.
By the way, here's a good rummage through the forum, found some interesting posts suggesting inputs
https://forum.mql4.com/ru/8835/page2 by plan
и
https://forum.mql4.com/ru/9321/page18#51761
2 Sart - if you are a beginner, you might be interested in the code from my post from https://forum.mql4.com/ru/12474/page9.
Preparing data for NS is not that difficult. There is another problem. Fat tails. I don't know how to handle it. I tried different variants. They get in the way quite a lot. Please ask Matemat about thick tails. I'll be resting for now. At least two weeks, but rest.
Preparing data for the NS is not that difficult. There is another problem. Fat tails. I don't know how to deal with it. I've tried different options. They get in the way quite a lot. Please ask Matemat about thick tails. I'll be resting for now. At least for a fortnight, but rest.
What do you mean thick tails, can you explain?
2 rip sorry i dont understand what it means. could you elaborate...
2 klot can you explain what it means?
2 Mathemat thank you very much, I remember...
What do thick tails mean. can you explain
That's a question for Math Math. I'm on holiday.
What does fat tails mean. can you explain
"Fat tails" is a term from risk management, defines a significantly changing price.
Yeah, I was just running through here. Fat tails is a fuzzy term from tervers. It's when the probability density function of a random variable's distribution decreases much more slowly than one would like when moving away from its expectation (if there is one).
Example: a Gaussian distribution, i.e. a bell-shaped curve. It has thin tails because exp(-x^2/2) is an extremely fast decreasing function. Hence the reasonability of talking about the three sigmas law: a value that is more than three standard deviations away from the centre of this distribution (here s.c.o. is equal to one) falls out in about 0.27% of cases. In other words, "large events" are rare and the vast majority of events fall within the "plus or minus 3 sigmas" interval.
An example of a distribution with thick tails: the Cauchy distribution (very similar to the finnings distribution, by the way). This distribution has a much slower decreasing probability density function, which approximates the law of inverse squares. Consequence: despite the fact that area under this curve is finite, there is no expectation and moreover no variance of this value (the corresponding integrals diverge in the usual sense). The point of talking about three sigmas is completely lost, as sigma itself simply does not exist. Large events have a much higher probability.
About financial series (in particular currency quotes): the distribution of closing price increments is a process in which the one-dimensional probability density function has thick tails. Therefore, indictors such as envelope, Bollinger bands, s.c.o., etc. do not make much sense. Catastrophes (collapses) are from the same series of events that people attribute a low probability to, but actually happen much more often. By the way, fat tails are very fond of breaking almost all traditional indictors: where we are expecting smooth behaviour of the dummy, it suddenly jumps to the unknown and gives a false signal.
The fat tails ("black swans") are very colourfully described by Taleb. There's a link, look for it here too.
Yeah, I was just running through here. Fat tails is a fuzzy term from tervers. It's when the probability density function of a random variable's distribution decreases much more slowly than one would like when moving away from its expectation (if there is one).
Example: a Gaussian distribution, i.e. a bell-shaped curve. It has thin tails because exp(-x^2/2) is an extremely fast decreasing function. Hence the reasonability of talking about the three sigmas law: a value that is more than three standard deviations away from the centre of this distribution (here s.c.o. is equal to one) falls out in about 0.27% of cases. In other words, "large events" are rare and the vast majority of events fall within the "plus or minus 3 sigmas" interval.
An example of a distribution with thick tails: the Cauchy distribution (very similar to the finnings distribution, by the way). This distribution has a much slower decreasing probability density function, which approximates the law of inverse squares. Consequence: despite the fact that area under this curve is finite, there is no expectation and moreover no variance of this value (the corresponding integrals diverge in the usual sense). The point of talking about three sigmas is completely lost, as sigma itself simply does not exist. Large events have a much higher probability.
About financial series (in particular currency quotes): the distribution of closing price increments is a process in which the one-dimensional probability density function has thick tails. Therefore, indictors such as envelope, Bollinger bands, s.c.o., etc. do not make much sense. Catastrophes (collapses) are from the same series of events that people attribute a low probability to, but actually happen much more often. By the way, fat tails are very fond of breaking almost all traditional indictors: where we are expecting smooth behaviour of the dummy, it suddenly jumps to the unknown and gives a false signal.
The fat tails ("black swans") are very colourfully described by Taleb. There is a link, look for it here too.
Wonderful! If I don't say it now, I probably never will. "Fat tails" are visible when we analyse quotes over 100 years and more... (exaggerated). If we take a specific section, e.g. the last 300 bars, there are no "fat tails" there... There will be a "tail", but when,??? Well, the hell with it, let it happen, but after the ejection the market will stabilize within Gauss again. So, if you take discrete plots, you can always fit into a system. The schemes for different parts of the market will be different....
Work to do, work to do....