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
The Grail seems to have been found.
It has yet to be proven in practice - but before I do, I take my hat off to you in public. Some clues have really helped me. Thank you.
Check on the history in the tester. MT5 has a tick history of real ticks.
To all lovers of the study of increments. Who is strong on this, is there a grain to this graph or is it pointless to move on?
And is it even possible to predict the next increment in this case?
1) Stationarity of increments is quite possible here
2) The increments are most likely dependent (there is a reversion after the outburst)
3) Using linear (and non-linear too) regression for prediction is problematic because the values of the series look discrete
4) We can try to use Markov chain instead of regression.
5) But the main thing is to make sure that the sequence is allowed to be modelled by a random process. This is not where the maths is of much help. For example, you could make a joke and lay out some deterministic sequence.
No, it's too early to say goodbye.
Here are the charts for EURUSD this week, with variance calculation using the formula D=(c*t*lambda)/4
And here is another one with the secret parameter
So, if we look at charts 2 and 3, this is the desired Grail. А?
So here we are again entering against the trend with the minimum amplitude of a backward move... Although it would have been more logical to enter the opposite way... That is, to move the whole theory from the left side to the right, hoping for the chance...
1) Stationarity of increments is quite possible here
2) Gradients are likely to be dependent (there are reversals after outliers)
3) Using linear (and non-linear too) regression for forecasting is problematic, because the values of the series look discrete
4) We can try to use Markov chain instead of regression.
5) But the main thing is to make sure that the sequence is allowed to be modelled by a random process. This is not where the maths is of much help. For example, you could make a joke and lay out some deterministic sequence.
no
no
Yes. This can be seen in the cumulative increment sum graph from the return.csv file in the attached archive:
Alexander, did you delete the post where you asked for a chart?
If not, I am attaching the archive with Mql4 code (maybe even 5 will work) and csv file.
Tell me if you need to change the formula or maybe I took it from the wrong place.
Yes. This can be seen in the cumulative increment sum graph from the return.csv file in the attached archive:
So if the increments are dependent, there's a chance? Took a live reading from the graph.Alexander, did you delete the post where you asked for a chart?
If not, I will attach the archive with Mql4 code (maybe even 5 will do) and csv file.
Tell me if you need to change the formula or maybe I took it from the wrong one.
No one is interested.
I need more lines of the confidence interval.
The formula is in private.
And no one is interested.
More confidence interval lines are needed.
The formula is in private.
The formula is better here.
read it
the formula is better here
read
From zero plus/minus 3*(SUM(ABS(return))/sqrt(240))