Our Masha! - page 10

 
or not - by changing the depth and therefore highlighting the harmonic, find frequencies with resonance and predict where the price will move
 

Unsteady harmonics in forex. It's not going to work:-(

 
about unsteadiness and redrawing:
it's not so bad if... looking for an MTS that would take the best qualities from a squirrel trader -
specifically would be the kind of MTS that would be reactive.
but this responsiveness is inherently the same overreaction!
 
Korey писал(а) >>
but this reaction in its essence is the very same overreaction!

Is it OK that you have to react with yesterday's number because of the redraw?

To clarify, you opened on a signal, and at the next count it was redrawn - reacting, so to speak, to the changing world:-) What do you suggest in this situation, to call the brokerage house and roll back everything as it was in the ass?

Here's the thing about our MASH.

I built a simple MTS that trades at its extremums opening in the direction of the sign of the first derivative of the MA:

The green color represents kotier (eurusd 1m), the black represents account balance without taking into account brokerage commissions. Everything is in points. It is clear that this shit manages to stick to the local trends and in general, even to be profitable.

 
Neutron писал(а) >>

...

I built a simple MTS that trades at its extremums, opening in the direction of the sign of the first derivative of the MA:

Green color represents kotier (eurusd 1m), black represents account balance without FC commission. Everything is in points. It is clear that this shit manages to stick to the local trends and in general, even to be profitable.

Here's the first tester I see on another platform. Very happy about it. And most importantly everything is correct in pips. The interesting thing is that it takes only half a day to create it on Matkadec and I have no problems with multicurrency.

Neutron

In version 14, you can use another Y-axis. It's very handy when you have different scales of the graph

>> you have a green curve, it's hard to see.

 

Thank you, Sergei, for the advice. I'll give it a try.

But I've heard from Grasn once that version 14 is glitchy in some places. Have you encountered this problem?

 
Neutron писал(а) >>

Thank you, Sergei, for the advice. I'll give it a try.

But I've heard from Grasn once that version 14 is glitchy in some places. You have not come across this problem?

Russificator not put. With him problems appear (the accuracy of the reflection after the decimal point is not regulated). Without it all normal works.

 

to Neutron, Prival


In 14 they replaced the solver with a very cheap one, it doesn't even really solve all the diffs... They had a fight with the previous supplier. Prival, this axis appeared a long time ago.

 
grasn писал(а) >>

to Neutron, Prival

In 14 they replaced the solver with a very cheap one, it doesn't even really solve all the diffs... They had a fight with the previous supplier. Prival, this axis has been around for a long time.

I've been using version 14 for about a year now. I seem to have solved the diffusers. I have them not so complex, maybe that's why I haven't encountered them. But thanks for the information. I had no idea. I will be careful now when solving diffusions. Although I have 1000 orders of magnitude more confidence in my macadam than in MQL. It rarely let me down.

 
Prival >> :

I've been using version 14 for about a year. I seem to have solved the diffusers. I don't have them complicated, maybe that's why I haven't bumped into them. But thanks for the info. I didn't know that. I will be careful now when solving diffusions. Although I have 1000 orders of magnitude more confidence in my macadam than in MQL. It has seldom let me down.

I've given somewhere on this forum a link to a detailed analysis of differences and bugs, and I'm not going to search for it now. Mathcad can't do many things, for example it can't solve trigonometric equations at all. The undeniable best in this class of programs is mathematica (http://www.wolfram.com/) . The only one better than mathematica is mountains. :о)


I would switch to mathematica but I am planning to use MT/MATHCAD/VISSIM. VISSIM is just a no-objection (http://www.vissim.com/), for those who have worked with similar systems. Here's a simple example of how to build a filter:




Either you can smooth something out




Or a fractal could be launched:




Or you could also run a satellite in real time:




Well, these are just examples :o))). Saves time, and works perfectly with MathCAD You don't need to transfer anything to MQL. You can integrate your own calculation blocks for modelling. I even wanted to write an article and open topic on this subject in MT/MATHCAD/VISSIM - but no time to do it. I have already prepared a lot of files, but this thing does not see the data from MT via DLL. All the time it generates "N/A", I've given up on this crap :o))). And then there are all sorts of integration issues that need to be solved :o(


I think I've solved diffuras. >> They are not complicated for me.

And I have enough for my primitive stochastic diffs :o)

Reason: