Machine learning in trading: theory, models, practice and algo-trading - page 761

 
Vizard_:

Come on Equi, let's have a laugh.

He who laughs last laughs.... You should post yours, I think it would be hilarious...

 
Combinator:
I would not buy it if I was a trader.

I think in the near future to open a UXO account, that's what I will lock... There everything will be seen ...

 
Mihail Marchukajtes:

He who laughs last laughs.... You put yours out there, I think it will be hilarious...

Teacher, all hope is up to you. Show me.

 
Vizard_:

Master, you're the man. Show me.

That's what I needed to prove. I have a minus, but you know how to throw poo professionally... Good luck!!!!

 
Mihail Marchukajtes:

That's what I needed to prove. I have a minus, but you know how to throw poo professionally... Good luck!!!!

Mishan, I'm tired of laughing, let's cry EQ)))
If it touches me, I'll tell you how to replace one of the inputs...

 
Vizard_:

Misha, I'm already tired of laughing, let's go to Equi, let's cry)))
If it touches you, I'll tell you how you can replace one of the inputs...

I'll post the trades today.... I think it will touch you, especially when you read the description...

 
Vizard_:

Let's do it with pictures. For example. Instead of z - your data.

---------------------------------

library(corrplot)

# generate the data

data <- seq(0, 2*pi, 0.1)
x1 <- 5*cos(data)
x2 <- 2*sin(data)
target <- ifelse(x1 + x2 > 0, 1, 0)

# let's assemble it to dataframe

z <- data.frame(data, target, x1, x2)

head(z) # See what happens
str(z) # + structure

# Let's look at the spread charts ("boxes with whiskers")

boxplot(z[,-1])

# And the correlation matrix

corm <- cor(z[,-1])
corrplot(corm, method = "color", addCoef.col = "darkgreen",
addgrid.col = "gray33", tl.col = "black")

It's a little unclear what this example is and what it's for, but I haven't redesigned my models yet, so I haven't made a training file either. When I'm going to retrain and make a trainee sample, I'll be sure to check.....

 
Mihail Marchukajtes:

It's a little unclear what this example is and what it's for, but I haven't redesigned my models yet, so I haven't made a training file either. As soon as I'm going to retrain and make a trainee sample, I'll be sure to check....

Misha, I'm fucking tired of laughing at you and Maximka the stoner.)

Show me.

# Let's see the spread charts ("boxes with whiskers")
# And the correlation matrix


On your data...

 
Vizard_:

Misha, fuck, I'm getting tired of laughing at you and Maximka the stoner)))

Show me -

# Let's see the swing charts ("boxes with whiskers")
# And the correlation matrix


On your data...

Yes I do not have them now, the training files from the current models I erased. When I will do retraining, then I will make your calculation and show everything. I'm not putting in all 110 predictors right now, but I'm selecting the ones that are relevant at the moment. I think I'll do something by the end of the week...

 

Here is the balance in two weeks, this is the third. The same model works. Last week there was a change of futures, so there were few trades, as soon as I changed the futures, it started working. Yesterday I caught two craps and immediately stopped trading. Today I restarted it in the evening, although I have missed one good signal. But on the whole, everything is good so far. I wish you to do the same.