Machine learning in trading: theory, models, practice and algo-trading - page 2982

[mytarmailS]

Renat Akhtyamov #:

congratulations ;)

one of my first tricks is revealed

as far as I remember, I substituted the transformed value of the volatility indicator instead of the period.

I think it was APAC

Congratulations ! this chip is called "adaptive filtering" "DSP" this chip is about 70 years old.

[Renat Akhtyamov]

mytarmailS #:

Congratulations ! This chip is called "adaptive filtering" "DSP" this chip is about 70 years old.

I'm late.

I already threw it away about 12 years ago.

Two MAs is a turtle's race.

I mean, it looks nice, but it's a hassle to use.

;)

[mytarmailS]

I came across this question from a Pythonist to R code.

I got a slight shock mixed with laughter...

this is how the problem is solved in python.

def to_supervised(train, n_input, n_out):
    X, y = list(), list()
    in_start = 0
    #  step over the entire history one time step at a time
    for _ in range(len(data)):
        #  define the end of the input sequence
        in_end = in_start + n_input
        out_end = in_end + n_out
        #  ensure we have enough data for this instance
        if out_end <= len(data):
            x_input = data[in_start:in_end, 0]
            x_input = x_input.reshape((len(x_input), 1))
            X.append(x_input)
            y.append(data[in_end:out_end, 0])
        #  move along one time step

        in_start += 1
    return array(X), array(y)

and this is how it's solved in R.

em <- embed(x = 1:10,dimension = 6)[,6:1]
e1 <- em[,1:3]
e2 <- em[,4:6]

So to say, feel the difference, which language is created for working with data, and which one just mows.

[Maxim Kuznetsov]

mytarmailS #:

I came across such a question from a pythonist to R code

I got a slight shock mixed with laughter.....

Here's how the problem is solved in python.

and this is how it's done in R

So to say, get a feel for the difference, which language is designed to work with data, and which one is just a mower

You're the one who didn't count strings in MQL :-) The language is designed to work with data

[mytarmailS]

Maxim Kuznetsov #:

you have not counted strings in MQL :-) Language for working with data

I don't even want to think about it ))

[Renat Fatkhullin]

Maxim Kuznetsov #:

you have not counted strings in MQL :-) Language for working with data

It's just you are not aware of the new functionality of standard matrix methods of MQL5:

• https://www.mql5.com/ru/docs/matrix

A big step has been made in basic matrix and vector maths. Writing can be significantly shorter.

[Maxim Dmitrievsky]

mytarmailS #:

I came across this question from a pythonist to code by

I got a slight shock mixed with laughter...

this is how the problem is solved in python.

and this is how it's done in R.

So to say, feel the difference, which language is designed to work with data.

The main thing is that you don't get a bitter aftertaste after laughing.

vec = [i for i in range(15)]    
matrix = [vec[i:i+7] for i in range(0, 14, 7)]
matrix2 = [vec[i:i+7] for i in range(1, 15, 7)]

It's even shorter with numpy. Do something meaningful.

[mytarmailS]

Maxim Dmitrievsky #:

Do something meaningful.

For example

[Maxim Dmitrievsky]

mytarmailS #:

Like.

I don't know, casual.

[mytarmailS]

Maxim Dmitrievsky #:
I don't know, casual.

What's that?