From theory to practice - page 668
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
Is 1.95 the variance there? It looks like standard deviation. The kurtosis coupled with the variance, the greater the variance, the less the kurtosis reacts to outliers, and vice versa, the lower the total variance, the greater the coefficient on outliers.
Nah, there's some internal variable there that's irrelevant.
As it is, yes. Oddly enough, it's the kurtosis (I calculate it with non-parametric metrics) that's responsible for the severity of the tails, which is what's required.
I don't want anything. Especially not from you.
That's fine. Empty pockets won't be long in coming.)
That's fine. Empty pockets won't be long in coming).
Sorry to intrude on your argument, if anything, but you give the impression of being an empty-nester. At least I've had that impression for the entire existence of this thread.
You make a clever face and sometimes ask such a question that I want to ask if it is not clear to 600 pages what a person counts, increment or the sum of increments.
But don't say again about the screenshot in the profile.
That's fine. Empty pockets won't be long in coming)
:))) So what the hell with them.
Bass, I'm speaking with more restraint - I (like others, I think) don't need questions, secrets, transparent allusions and other crap.
I want answers, literature, graphs, figures - i.e. a professional approach to the matter. If you can't offer that, get out of the thread.
You make a clever face and sometimes ask questions that make you want to ask if it's not clear by page 600 what the person is counting, the increment or the sum of the increments.
Questions are a form of asking the author to think. I'm not interested in the answers, I already know them.
Regarding the correlation - if the author meant the increments, their correlation is very weak and does not "guarantee" anything, as he wrote.
And if he meant the sum of graphes, their negative correlation is stronger, of course, but we don't trade the sum of graphes, i.e. the price with deleted trend, but the initial price with trends present. The negative correlation of the sum of graphes "guarantees" profit only on the graphes chart, but not on the price chart.
Therefore the ACF of the sum of increments is self-defeating. Some kind of trend detector has to be attached to it as well.
Therefore the ACF of the sum of increments is self-defeating. Some kind of trend detector needs to be attached to it.
Alexander, I really need your help in processing this data (attached).
What is the histogram there and are there any statistical patterns or not. Quantiles, etc. what can you say.
Because I'm a sucker for this stuff.
The interesting thing is that in the post above I used a multiplier of 1.6, a strange coincidence, but it was Chegevara all the time mentioned this number.
Alexander, I really need your help in processing this data (attached).
What is the histogram there and are there any statistical patterns or not. Quantiles, etc. what can you say.
Because I'm an amateur in this business.
What's interesting is that in the post above I used a multiplier of 1.6, a strange coincidence, but it was Chegevara who kept mentioning this number.
If you remove 0, of which there are VERY many, it looks like this.
No, well, I can't tell right away what you can get out of it. Depends on what goal you're aiming for...
About EURUSD 2018...
If one considers the non-parametric kurtosis of the sum of increments relative to the median, and sets a kurtosis entry condition <10, then the devastating trades of June 14 and September (see the charts highlighted in red circles) would absolutely fail.
Can I ask for a histogram of this amount. Maybe there's something interesting there.