Machine learning in trading: theory, models, practice and algo-trading - page 2922
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
If you look at the history of unchallenged posts and what was going on there in general, you get a bad impression that the provocateurs got away with a scare and after a pause are actively spamming the thread. The last few pages are so directly led by these "experts in ML/DL/NN".
Okay. Calm down.
He was banned for unseemly insulting the moderators. To put it mildly. It's probably gonna be a long time.
Oh, ONNX is here. We'll live better than ever now.)
ChatGPT and NLP are next)
Well, ChatGPT is possible via the OpenAI API, and NLP is quite possible via ONNX.
The TS is based on predicting the next bar on R and then using this prediction in an Expert Advisor on MKL4.
Model on R.
It works on H1, the teacher is trending (ZZ), predicts the next bar. OutOfSampe is not used, as the model is recalculated at each bar.
Efficiency of predicting the next bar 78-80%
Positive prediction performance on one of the next 8 bars is over 95%.
The model obtained is excellent.
BUT.
It is impossible to build a working TS on this model. We get about 78% of profitable trades, but they are small. But the losses are much larger.
In the EA itself we cut losses and let profits grow. This leads to the fact that the number of profitable trades decreases with simultaneous growth of a profitable trade and reduction of a losing one. But it still doesn't solve the problem.
Here is one of the many results of the TR and SL exercise.
If we look at the chart with deals, we can see that the erroneous predictions fall on strong market movements, i.e. the model for some reason erroneously predicts the directions of strong market movements. The reasons are not clear, as the model itself is trained on the sample obtained by Sample.
Typical chart.
It is hard to see, I have highlighted deals with vertical lines
Well, ChatGPT is possible via the OpenAI API, and NLP is quite possible via ONNX.
API GPT is good, but the black box GPT is not kosher to manage money))))) NLP seems to be better. Or isolate the logic of the GPT))))
In Rennesans by the way by 2001 or 2nd year 300 thousand trades were made per day on 8000 thousand instruments, and it was not considered high-frequency trading, as most of the trades were split large trades that would not have a big impact on the market).
I have a few questions
It works on H1, the teacher is trending (ZZ), predicts the next bar. OutOfSampe is not used because the model is recalculated on each bar.
if it predicts the next bar, why does ZZ do it?
The efficiency of positive prediction on one of the next 8 bars is over 95%.
what do you mean 8 bars???, if the model predicts only one bar - namely the next one.
I have a few questions
if the next bar is predicted, what is the ZZ for that?
what does 8 bars mean???, if the model predicts only one bar, namely the next one.
for example, that 1 out of 8 is white :-)
There are a lot of strange things there in general - the model is being tested, but a parabolic and AMA are also included in the robot. There are strong suspicions that if you remove R and other stuff, the result will not change fundamentally - parabolic and ama in a pair will give similar results
I have a few questions
if the next bar is predicted, what is the ZZ for that?
what does 8 bars mean???, if the model predicts only one bar, namely the next one.
Why is not up for debate. This teacher is trending, hence the 8 bars, i.e. it picks up trends.