Machine learning in trading: theory, models, practice and algo-trading - page 719
![MQL5 - Language of trade strategies built-in the MetaTrader 5 client terminal](https://c.mql5.com/i/registerlandings/logo-2.png)
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
Unfortunately, no. I train models every week. My signal statistics are depressing, because there is a huge difference between tests and real trading. Not in the divergence of the results, but in the nuances of real-time.
I don't know why I got errors when I tried to rewrite it.
If I wanted to rewrite it for one bar but it will not work for the whole period, it should be recalculated for the last 100 bars.
Unfortunately, no. I train models every week. My signal statistics are depressing because there is a huge difference between tests and real trading.
Your model is overtrained, it only understands what it was trained on.
The error on all three parts of the first file should be the same.
After that, check again on the second file. This file should be perfectly normal with increasing dates
Again the error should match the three errors in the first file.
If all four errors differ by more than 5%, then there is no model.
Hi!
Guys, when will the super bot with AI finish?
You'll get old waiting.)
NEVER.
I do not sympathize with his teachings and resent being compared to him. I agree there is a certain lack of clarity in the article itself, or something like that...... But here are the specifics. I adhere to the following model of cause and effect.
First, a market expectation is formed (the volatility smile on the selected instrument. See the video above.) Then the delta volume and OM is traded in accordance with this expectation or not. Then comes the change of the price itself and then the change of ALL indicators drawn from the price. And only in this order, not the other way around......... So when you try to build a price forecast based on an indicator. Super duper smart tricky machine, then you are initially breaking the cause and effect relationship. Hence the result.....
I wasn't talking about your article, I was characterizing the post:
Mihail Marchukajtes:
You see, what's important here is the approach to the market as a whole. If it's correct, the TS will not make you wait long. Read my article..... its main point is to show the approach to the market..... If you approach it correctly, you can get quite an interesting advantage. And those who agoldelo throw themselves on kotir as on BP unsteady...... They are usually left with what they had in the beginning. You have to take a broader look at the market. See what options are, how expectations are formed, how they are subsequently traded and how the price reacts after that. This seems to me to be one of the most correct approaches.... IMHO
This article has some specifics, but IMHO it is very primitive, like Safin's or Larry Williams'.
At the expense of the "volatility smile", OM and different deltas, it's all known for a long time. Taking into account the decentralization of Forex, as well as of the world's stock exchanges and limited access to relevant data, all this data, publicly available or under $1k a month, is squeezed dry or even a fake copy of the DC Forex volume.
Fa, look for interest in the influence of weather conditions of the cities of Pendostan (New York..,
Chicago...) and the EU the previous day, on today's eurusd candle color...
https://www.wunderground.com/history/airport/KORD/2000/1/1/CustomHistory.html?dayend=1&monthend=12&yearend=2000&req_city=&req_state=&req_statename=&reqdb.zip=&reqdb.magic=&reqdb.wmo=
An interesting point. The weather exactly in Chicago, where the exchange is traded, may affect the trading style of MM on that day. Well it's so.... Just thinking aloud...
I wasn't talking about your article, I was characterizing the post:
The article has specifics, but IMHO very primitive, like Safin or Larry Williams.
At the expense of the "volatility smile", OM and different deltas, it's all known for a long time, and given the decentralization of forex, as in fact the world stock exchanges and limited access to relevant data, all these data, which are in the public domain or at a price below $ 1K per month, squeezed dry or even fake as a volume in the DC forex.
No way. I use them and there are fish, maybe there are others, more effective to market data and for more expensive dough. But what is there, is there. Again, it all depends on how you prepare them, I mean OI deltas and volumes. Many even having access to them, make mistakes when preparing training.....
Your model is retrained, it only understands what it was trained on.
The error on all three parts of the first file should be the same.
After that, check again on the second file. This file should be perfectly normal with increasing dates
Again the error should match the three errors in the first file.
If all four errors differ by more than 5%, there is no model.
That's a cool plan, but I do it differently. I leave the OOS period and start building models while applying a certain approach to preparation, sifting, control analysis. And if the model on the OOS shows a satisfactory variant, then I believe that these manipulations in the preparation of the model are correct. I subsequently apply them during model building for trading.
Not the stock exchange, but the exchanges + cities of major banks and the Central Bank)))
Well yes, yes.... If their weather is bad, it's bad for all Chicagoans
Cool plan, but I do it differently. I leave a period of OOS and begin to build models while applying a certain approach to preparation, sifting, control analysis. And if the model on the OOS shows a satisfactory variant, then I believe that these manipulations in the preparation of the model are correct. I subsequently apply them when building a model for trading.
You are not confused by the result?
And you are not embarrassed by the result?
In the light of recent discoveries, it makes me very happy. All that remains is to transfer this joy to the bill. A picture of the oil:
Results in the tester since 23.02
And here is the real trading for the same period.....
So tests and real trading are completely different things....