From theory to practice - page 541
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 think I've figured out which function to regress to...
(If you don't remember, it's this one: #5220, #5222 )
You could try using two polynomials.
That is, 2 polynomials that are combined at a common point. The last point of the first polynomial is the first point of the second polynomial.
This thing can handle both this shape
and this
and this
and many others.
...
well, right
why?
hopefully not because the turkey drew it that way, whatever it is....Yes and really interesting, ejection is a skew, if it is completely extinguished, then the price will go as before, and if it is not completely extinguished, then the remaining energy will guide it towards the skew until it runs out, who else thinks so?
It's really interesting, the outlier is a skew, if it's completely extinguished then the price will go the way it did before, and if it's not completely extinguished then the remaining energy will steer it towards the skew until it runs out, who else thinks so?
A picture will tell you more than I do.
If the chart of any currency pair has a lot of such impulses - then this model is not suitable for such a currency pair.
And if they will be 50/50, we are talking about a long period of time, such phenomena on all currency pairs, and if the model does not take into account all variants, then how well the model will be adequate to the market, then it will be a model for the sake of the model.
And if they are 50/50, we are talking about a long period of time, such phenomena on all currency pairs, and if the model does not take into account all options, then how adequate the model will be to the market, then it will be a model for the sake of the model.
on all of them. But some more than others, some less.
Some of them can be killed by not trading the news. some of them you have to put up with, "as long as there are more good trades than bad ones".
And if the model does not account for all variations, then how adequate the model will be to the market, then it will be a model for the sake of the model.
if only more than 50% would count.
I already suggested to Georgi Merts not to discuss it. there's "defining channel boundaries", and we're mostly talking about "average" here.
Tried to implement this thing in excel.
Let the formula for the first polynomial: y=ax2+bx+c
the formula for the second polynomial: y=mx2+nx+k
we need to find coefficients a,b,c for the first equation and coefficients m,n,k for the second equation.
and X is also the coordinate on x-axis where the end of the first polynomial and the beginning of the second one will be.
i.e. there will be 7 unknowns.
In the attached table the coefficients a,b,c are shown in purple. m,n,k are in pink. X (big x) is shown in gray.
I have given the following formula in cell C2:
If the x-axis coordinate value (orange column) is smaller than X (X is large), we build a parabola by the formula y=ax2+bx+c,
. If the x-axis coordinate value is already larger than X , we build a parabola by the formula y=mx2+nx+k.
The Solution Finder tool refuses to pick up the value of X correctly.
What could be the problem?
You may try to try and enumerate...
Try values of X from 0 to 201. At each trial, count the sum of squares of deviations by two polynomials. At each trial this sum will be the smallest - that's what you should use.
I'd like to see what other disadvantages this approach has, and try to remove them.
i think that this system is better to trade on cointegrated pairs (eurchf, audnzd). people use a wand for this purpose. but the wand model is a horizontal straight line, while channels may follow different trajectories.
Try to implement this theme in mql, and post it in a branch (or at least in my personal message, if the grail works :)). and we'll see what other drawbacks this approach has, and try to remove them.
This system is better to trade on cointegrated pairs (eurchf, audnzd). people use the dash for this purpose. but the dash model is a horizontal straight line, and channels may follow different trajectories. therefore, you get either minimum profit or no profit.
The currency market is balanced, unlike the stock market. We have to dance from there. No indicator will work better than a wand. It lags, but it has other advantages.
I think I've figured out which function to regress on...
(if you don't remember, it's this one: #5220, #5222 )
You could try using two polynomials.
That is, 2 polynomials that are combined at a common point. The last point of the first polynomial is the first point of the second polynomial.
This thing can handle both such a shape
and this
and this
and many others.
...
MNC can handle ALL of these shapes.
I still don't understand why you have to make something up when it's all been made up a long time ago.
The MNC can handle ALL of these figures.
I still don't understand why you have to come up with something when everything was invented a long time ago.