Machine learning in trading: theory, models, practice and algo-trading - page 752
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
Michael is just bragging about the grail. And the rest of us create the effect that we can do it too :)
Doc, when are you going to put the amount of increments on one input and the trading intensity on the other and show everyone here the result of 80-90% like, sleeping for a month, Warlock?
I don't have trading intensity, but if the tick volumes that the kitchen will give me help - I will definitely write it here.
No, you can't do without them. This is one of the keys to the Grail. Or rather, the picklock.
Today I remembered one more transformation and decided to include it in the standard set and now I have 154 predictors. This is the standard deviation. In NS times this particular transformation showed the best results on classificatory networks..... Included it, let's see now...
Micha, you're about to reveal the whole set...)
Exactly right, It is necessary to have a model that as a person made the trade. A human trader
1) Makes a forecast
2) Assesses the risks
3) Meeting the risks of entering the trade
4) Exit from the trade
Everything approximately, an outline as they say)))))
Suppose, I have laid on indicators, including those for different TFs, traded it by hand and got financial results. There are entry points, there are indicators, there is a financial result, both positive and negative. Why not teach network neurons on this material?
And yet, maybe the network can say that when the event X occurs, we always close - give an unconditional instruction that will be a trawl, and besides inform the vector that should be used to try to enter, i.e. break through the level (detected by the indicator) - there is a signal to enter, and the network thinks, based on indicator/pattern indications, whether it is worth it to enter or not.
Micha, you're about to reveal the whole set...)
Unfortunately, the Standard Deviation did not show itself this time.....
For example, let's say that I have laid out the indicators, including on different TFs, traded it with my hands and got the financial result. There are entry points, there are indicators, and there is a financial result, both positive and negative. Why not teach network neurons on this material?
And yet, maybe you can tell the network that, when the event X occurs, we always close - give an unconditional instruction, which will be a trawl, and in addition inform the vector that is worth trying to enter, i.e. break a level (detected by an indicator) - there is a signal to enter, and the network thinks, based on indicator/pattern readings, whether to enter or not.
You are right, comrade!!!! First, we determine the event X, in which we plan to analyze. Once event X has occurred, feed a set of indicators to the network input and get a yes to trade up, no to trade down response from it. This is the basis of any strategy using the AI.... The correct use of AI.
Why bother with neural networks? There is a system and rules. The system brings results. What prevents it from being programmed? In my opinion, the point of NS is to find regularities.
For example?