Machine learning in trading: theory, models, practice and algo-trading - page 666

New comment
 

Gentlemen, is the derivative:

from M1 is M5,

...

and from W1 it's MN.

?

If this is true, then why do we need these NS, because the dependence already exists ...

why look for it?

 
Renat Akhtyamov:

Gentlemen, is the derivative:

from M1 is M5,

and from W1 it's MN.

?

If this is true, then why do these NS, because the dependence already exists?

why do you need to find it?

The derivative(of a function at a point) is a basic concept of differential calculus, which characterizes the rate of change of a function (at a given point). It is defined as the limit of the ratio of the increment of a function to the increment of its...


M5 is not a speed from M1.

 
elibrarius:
Derivative(functions at a point) - the basic concept of differential calculus, which characterizes the rate of change of a function (at a given point). It is defined as the limit of the ratio of the increment of a function to the increment of its...

Derivative of a function - Wikipedia

M5 is not the rate of change from M1.

fine

OK, let's have an integral.

M1 rolls up into M5, .

and yet?

 
Renat Akhtyamov:

Well, that's fine.

All right, let's call it an integral.

and yet?

In integrals, everything is blurred and delayed. I.e. dependencies are lost.
 
elibrarius:
Everything is blurred and lags in integrals. I.e. the dependences are lost.

Let's assume we have a last-order convolution - a single candle MN1

respectively, below the middle we will buy and above the middle we will sell

Try to predict the sequence of actions of a price-quoter who knows that the price will fall below the current low of the candle - he has already won, he does not need to do anything.

the same is true for the NB, i.e. it will fail anyway, no matter how hard you try.

 
Vizard_:

My Teacher is the best!))


getting our pockets ready, dusting them off

 
Maxim Dmitrievsky:

Preparing our pockets, dusting them off.

He's making fun of me.

Get your valerian ready

Forum on trading, automated trading systems and trading strategies testing

Machine Learning in Trading: Theory and Practice (Trading and Beyond)

Vizard_, 2018.02.11 09:00

And you thought you were the only one getting stoned)))) hilarious...

According to the drawing kuyna builds. If the "tomorrow is like yesterday" principle works, then fine, otherwise...

It's time to rename this thread "Let's smoke one")).


 
Vizard_:

Any model is a formula. You roll it up in a dll and plug it in anywhere.

Another option is scalable graphs in R http://jkunst.com/highcharter/

I'd roll them right into MT. And to the UPU.

No one has found a converter to code?

 
Renat Akhtyamov:

He's mocking.

take valerian


I wonder about the convolution, by the way... is there any way to apply it to understand which of the 2 signals is ahead and which is lagging?

 
Maxim Dmitrievsky:

I wonder about the convolution, by the way... is there any way to apply it to understand which of the 2 signals is leading and which is lagging?

Volchanskiy specializes in DSP

Try asking him how he does the convolution with a MCU

the lag is usually the most averaged one
1...659660661662663664665666667668669670671672673...3495
New comment
Reason: