Machine learning in trading: theory, models, practice and algo-trading - page 1845
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
And most importantly all at once here, for evaluation, so to speak, by independent experts..... I know we are pulling the branch together, but what about the corners? :-)
Max, what happened to your laziness? Tell me how you conquered it.
What are you talking about? Are you drunk again? On Monday morning? :))
What are you talking about? Are you drunk again? On Monday morning? :))
Nah, I'm sober as a whistle.... I'm just in a good mood. I'm going to go vote, to show my civic responsibility, so to speak!!!!!
You know, you use so many dithyrambs in your speech that I find it difficult to understand you. Write more simply. We are all dummies here. I, for example, Maxim, Maxim there Max is also a teapot :-) I am also a troll :-)
I think I wrote as simply as possible ))
But never mind, I'll try again.
The price you generalize only by timeframe. Do you have any idea what else to generalize the price to?
Endogenous data is everything that refers to the numerical internal data of the model, in our case it's anything. Quotes, volume, time, its attributes, etc.
I simply call this data technical.
Exogenous data is anything that refers to external data, but which affects internal data. News, reports, events, etc.
I simply call this data fundamental.
I thought it was understood here ))
It seems to be as simple as possible ))
But never mind, I'll try again.
You generalize the price by timeframes only. Do you have any idea what else to generalize the price to?
Endogenous data is everything that refers to the numerical internal data of the model, in our case it's anything. Quotes, volume, time, its attributes, etc.
I simply call this data technical.
Exogenous data is anything that refers to external data, but which affects internal data. News, reports, events, etc.
I simply call this data fundamental.
It seemed to me that it is understood here ))
There is a suggestion to combine efforts ... As you have seen my entries are good but not many, so I need to trade all the tools at once, I need a robot, since entries are often 90% at the very minimum, I need to enter quickly to keep the advantage, I need a robot.
Karoch. there is a proposal - to create together ATC.
1) I will generate the TS in my R . TC in the form of a file with log. rules.
2) You make a tool that will open trades in MT4 or MT5 on these rules.
What do you think?
I don't know, I'm lazy.) I have a bot that trains properly, i have screenshots of it... i don't trade with it... i'm too lazy
And most importantly all at once here, for evaluation, so to speak, by independent experts..... I know we are pulling the branch together, but what about the corners? :-)
Max, what happened to your laziness? Tell me how you conquered it.
I caught 2 perch by the kilo yesterday, swam in the boat, sunburnt, shaloputnichal ))
By the way, where do you get the history of OI?It seems to be as simple as possible ))
But never mind, I'll try again.
You generalize the price only by timeframes. Do you have any idea what else to generalize the price to?
Endogenous data is everything that refers to the numerical internal data of the model, in our case it's anything. Quotes, volume, time, its attributes, etc.
I simply call this data technical.
Exogenous data is anything that refers to external data, but which affects internal data. News, reports, events, etc.
I simply call this data fundamental.
It seemed to me that it is understood here )))
Judging by your description, you call a model the market for a particular instrument/currency. That is, the parameters that describe the price.
A model is usually called an MO program (a neuronet, a forest, drilling), and more exactly - the structure of weights, coefficients, links between elements built by this program. Which may have a lot of customization parameters (a share of rows, a share of columns, depth - for a forest; number of layers, number of neurons in a layer, activation functions - for neural networks, and this list is not complete), these parameters are called hyperparameters of the model. At first I thought you meant them as endogenous (internal).
No one here has worked with fundamental data. Some people wanted to train the model using MT5 news but they say it cannot be read in the tester. Although, it is possible to save everything in a file and use it later.
This is a task for large companies, where programmers and traders work for money. Here everybody is experimenting on his own, spending his time only on what he considers promising. It's not promising to analyze the events alone, because it's a task for thousands of hours of work with no clear prospects.
Recently, there was a link to a company that wants to trade by analyzing twitter.
Caught 2 bass a pound yesterday, swam on the boat, burned in the sun, shaloputnichal ))
I've been suggested to try to build a robot on it.No history, collected on the VPS, but now abandoned it. I've never managed to compile it, but I've never managed to make a reliable indicator for using it in my EA. The quality increases by 20-30%. I judge by the number of selected inputs. If I do not have OM, then 100-120, but as soon as I add OM to the file, it immediately selects 150-180 significant entrances where OM is not in the last place. But I cannot use it automatically. I have an Expert Advisor that successfully saves it in a file. But the indicator that builds it looks ok, but when I feed it to my EA it gets shifted by one bar and I get really messed up. If I can change it I would be glad.
Judging by the description, you call a model the market of a particular instrument/currency. I.e. parameters describing the price.
In this case the model is usually called the MO program itself (neural network, forest, boosting) which may have a lot of customization parameters (share of rows, share of columns, depth - for forests; number of layers, number of neurons in a layer, activation functions - for neural networks and this list is far from complete), these parameters are called hyperparameters of the model. At first I thought you meant them as endogenous (internal).
No one here has worked with the fundamental data. Some people wanted to train the model on the news, but they say that they cannot be read in the tester. Although it is possible to save everything in a file and read it later.
In contrast, analyzing the news/events is not a task for individuals, but for large companies with programmers and traders working for money. Everyone is experimenting on his own, spending his time only on what he considers promising. It's not promising to analyze the events alone, because it's a task for thousands of hours of work with no clear prospects.
Recently there was a link to a company that wants to trade by analyzing twitter.
Thank you for your comprehensive reply. You got my point across correctly.
Yes, that's what I was wondering about the fundamentals. But, as you write, no one has worked with it because it requires knowledge of the fundamental analysis.
The analysis of social networks, yes, it's a good idea.
The Americans also write about it, they see this as a promising future in the development of MO.
Thank you for your comprehensive reply. You've made a good point.
Yes, this is what I was wondering about the foundation. But as you write, no one has worked with it, because it requires knowledge of the fundamental analysis.
The analysis of social networks, yes, it's a good idea.
They see it as a promising future in the development of IO.