Machine learning in trading: theory, models, practice and algo-trading - page 2981
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
minimum per bar?
it's 5 minutes, trades +-5 pcs a day, what spread, calm down.
If you control the periods of the mashes in dynamics, you get a good TC on the mashes.
Moreover, you can try to predict which periods of the mashes will be profitable in the future.
Interesting.
Are you using a regression model? How did you lay out the target?
Interesting.
Are you running it with a regression model? How did you mark up the target?
The model is not regression model, but you can make regression model, there is no target model, selection through AO.
the model is not rkrxion, but it is possible to make a regression model, there is no target model, selection through AO.
What do you mean by "selection" - is it random/genetics or is there feedback through what function?
There is a basic strategy on MA, and to work in the best light we look for some coefficients that correspond to the MA period, right?
1) What does "selection" mean - is it random/genetics or is there feedback via what function?
2) There is a basic strategy on MA, and some coefficients responsible for the MA period are looked for to perform at their best, right?
h ttps:// youtu.be/cMDpWf-aSrs
The usual propaganda.
Europe only after 1500 years has almost reached (except for sanitation) the standard of living that was in the Roman Empire.
The whole dog is buried in the methodology of calculating GDP, which is divided extremely unevenly per capita (today up to 10 million people die of starvation every year). Today the other trouble is that GDP includes financial and other "paper" assets - there are no GDP figures in physical terms.
Rubbish.
The growth of scientific discoveries that gave rise to technological progress - yes, and this has led to explosive growth in labour productivity, allowing us to feed 8 billion people. But if you compare the middle of the population, setting aside the 3% richest and 3% poorest, the Roman Empire, which together with Byzantium existed for 1500 years, lived very well despite the lack of iPhones and the internet.
If we set aside the blatant propaganda, the most interesting thing in the video for this thread is that predictions are practically useless.
The whole dog is buried in the methodology of calculating GDP, which is divided extremely unevenly per capita (today up to 10 million people die of hunger every year). Today there is another problem: GDP includes financial and other "paper" assets - there are no GDP figures in physical terms.
Rubbish.
I support it completely. It is always irritating when all sorts of "economists" with a serious face rant about GDP and the values of its "growth" (the nature of which is largely in the inflation of money), especially if we are talking about fractions of per cent. Even if there were a normal and uniform method of calculation, indicators like GDP would still be meaningless, because they are measured in money, the purchasing power of which varies over time. It's like trying to measure something with rulers that are constantly changing size. If we are going to measure the economy, which is about the production and consumption of goods, it is the quantity of these goods.
A beginner's guide to neural networks in python.
If you control the periods of the mashes in dynamics, you get a good TC on the mashes.
Moreover, you can try to predict which periods of the mashes will be profitable in the future.
trying different variations of fitness functions
congratulations ;)
one of my first tricks has been solved
as far as I remember, I used to substitute the transformed value of the volatility indicator instead of the period.
I think it was the APR