Machine learning in trading: theory, models, practice and algo-trading - page 2021
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
Well, then take it and trade the first two days and do not fuck your brains out ))
or try to pre-train online
This week it works till Thursday ) but there has been no market reversal yet. It is unclear how quickly will respond
i dont know how to use lstm ? try it, it's cool
This week till Thursday is holding on, but there have been no market reversals yet. It is unclear how quickly the market will react
Do you use lstm? Try it, it's cool
Only Sell?
Only Sell?
Where to buy there?
Where do you buy there?
You train only Sell or both directions, and the grid works itself?)
You train only Sell or both directions, and the grid works itself?)
by myself
This week till Thursday is holding on, but there have been no market reversals yet. It is unclear how quickly the market will react
Have you used lstm? Try it, it's cool
i.e. is it like a decision cycle?
This week till Thursday is holding on, but there have been no market reversals yet. It is unclear how quickly the market will react
Have you used lstm? Try it, it's cool
why does neuro open a position every 6 candles?
i.e. is it like a decision cycle?
not every, it just happens
I decided to make an analogue of the tester, not to run training and its evaluation in the terminal tester, but for NS/forest to quickly evaluate the results of training.
I plan to take as input data for calculations:
1) OHLC by Bid
2) commission - can be charged at the moment of trade opening
3) swap - to be charged at the moment of day change
4) OHLC by Asc - instead of spread. Since the spread on the bar is specified as the minimum for the lifetime of the bar. At the moments of OHLC spreads were most likely not minimal, but larger. During sharp movements, they were much larger. It will be necessary to take it from real tick data, i.e. it will require the complete walkthrough of real ticks.
A request to developers (if they will look here) - may I add OHLC on Asc to information about bars?
Is there something else to consider?
Maybe there is a ready-made tester? Even without OHLC on Asc will do as a basis for refinement.
And if it is pending, then it is necessary to skip deals at very high minute candlesticks (gaps).