From theory to practice - page 890
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
I've read books that you couldn't even dream of, Yuri. Though you're obviously better in your field than I am - I'll give you that.
A bit of philosophy, though.
The more I understand these equations, methods of collecting quotes, dispersion calculations, the more I am amazed at the amazing beauty and accuracy of structures in the market. The price-time space is inextricable. The slightest inaccuracy in the calculations leads to a scattering of results of 100% or more. Each component of the system: the type of distribution, the measure of central tendency, and the variance are crucial and neglecting any of them is unacceptable.
Nah, beautiful in the market, gentlemen. Fascinating.
I've read books that you wouldn't even dream of, Yuri.
Thank God for that. Nightmares are all I need.)
Bit of philosophy, though.
...
Nah, it's beautiful in the market, gentlemen. Fascinating.
In general, you will not find a substitute for MA, from the word - never. It just doesn't exist in nature, so don't look for it. In short, you're sitting on long tails. Either work with regression, or change the conditions of the problem. There's not much choice.)
In general, you will never find a substitute for MA. It simply does not exist in nature, you don't have to look for it. In short, you are sitting on long tails. Either work with regression, or change the conditions of the problem. There's not much of a choice.)
I've done enough research to know that the SMA is not a bad choice. However, among the measures of central tendency (see Wikipedia) there are definitely things better than it. The results are bluntly improved. And let everyone decide for themselves whether it's worth working on. I no longer intend to expose my research to lazybones like FelixWhite and the like.
I have done enough research to know that SMA is not the worst choice.
SMA is the worst choice. WMA is one of the best, but calculating the coefficients is a real challenge. EMA and its modifications are intermediate in terms of lousiness.
SMA is the worst choice. WMA is one of the best, but the ratios are a challenge. EMA and its modifications are intermediate by degree of lousiness.
:))) You're smart and experienced, my friend.
I have done enough research to know that the SMA is not the worst choice. However, among the central tendency measures (see Wikipedia) there are clearly things better than it. The results are bluntly improved. And let everyone decide for themselves whether it's worth working on. I no longer intend to expose my research to lazybones like FelixWhite and the like.
Actually, there are two sma. Which one has the latest value/period at the output of the window and as in the rsi calculation, and these are two big differences.
Mascots are a good choice. As cheesy as it sounds, MAs can be had in many ways :-) MA as a HF filter, MA for comparing with the past on its reference period, MA for following the trend.
It's a great indicator because it shows a lot of things at once, it's different for every TS.
---
how do you want to get the normal distribution of the difference with MA without getting rid of its periodicity and moving window errors ?? You should either look for some aperiodic averaging (I can't even imagine that) or get a technique to eliminate periodicity-induced errors
I have read such books which .........
Did someone create such books....
Therefore, theory can be read as well as created.
It is clear that all the existing theory on the market does not work, as no one has been able to make money since the 70s.
There is no point in reading these books, create new formulas, create in short.
Can they?
Has anyone created such books....
Therefore, theory can be read as well as created.
It is clear that all the existing theory on the market does not work, as no one has been able to make money since the 70s.
There is no point in reading these books, create new formulas, create in short.
Are they able to?
Here's one about asymmetries and excesses
https://www.mql5.com/ru/articles/273
and here's a test for normality. The Harky-Bera test.
https://www.mql5.com/ru/articles/257
In fact, you can come up with your own measure of normality.
For example: multiply [asymmetry coefficient] by [measure of deviation of kurtosis coefficient from normal kurtosis coefficient].