Machine learning in trading: theory, models, practice and algo-trading - page 467
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
Here is an example, the same data, but using only arima. Predicted the last 1000 values from the first 5712.
Here is an example, the same data, but using only arima. Predicted the last 1000 values from the first 5712.
Code and csv in atach.
Respectfully.
I uploaded the slightly corrected r script, there are now a few more estimates of the model, and a couple of errors corrected.
r^2 on training data: 0.8742684
r^2 on the test data: 0.8150772
r^2 on test data, calculated for price increase (diff): -0.8990404 (arima predicts very large studs, though the direction in general is correct, but the price increase forecast is much larger than necessary, r^2 is completely spoiled by this)
accuracy of price growth direction prediction on test data: 0.7257257
I uploaded the slightly corrected r script, there are now a few more estimates of the model, and a couple of errors corrected.
r^2 on training data: 0.8742684
r^2 on the test data: 0.8150772
r^2 on test data, calculated for price increase (diff): -0.8990404 (arima predicts very large studs, though the direction in general is correct, but the price increase forecast is much larger than necessary, r^2 is completely spoiled by this)
Accuracy of price direction prediction based on test data: 0.7257257
Regards.
The residuals are autocorrelated, which is not good. Well you can see that for yourself...
OOO And here comes our all-time favorite Trickster!!!! Got your hands full of poo???? Here's a reason for you to throw them around....
On the subject of whether Reshetova's optimizer is retrained or not. This is the screenshot I posted yesterday in one of the forex groups..... Blue shows the optimization period, green - futures contracts. But there is no possibility to submit the data and it turns out that since the beginning of the year it worked fine.... And you say overtraining, you just need to be able to train....
Now what are you going to say about overtraining????
Maybe you will increase accuracy of prediction.
Regards.
Residuals are autocorrelated, which is not good. You can see it yourself...
If you compare the predictions in the article and these, you can see that the predicted trends coincide perfectly in both models, but in the article the model is much better at catching sharp jumps. And the arima - with price spikes as luck would have it, and those "bad luck" will cause the biggest drawdowns. Plus in the properties of the model it is clear that seasonality is not used. For now the article wins :(
It still takes a lot of Arim's intuition to correctly set the search limits for ar,i,ma coefficients, and make the model look for seasonal parameters.
I turn to the professionals... Who really chip in MKUL4 not to ask questions or as Klot said, the devil is programmed ... That's exactly what I need.
The advisor should be guaranteed to open and close trades in accordance with the logic. For a knowledgeable person this is a three-hour job in realtime mode... ... we will agree and do it. There it is written well, really not difficult and fast, just need a RELIABLE execution with handling all server errors, with pauses, etc. This will be discussed in the process of..... In turn, I start this EA on VPS and signals copying will be organized. Accordingly, I can connect to the project..... Maxim, tocxic, forexman77
I turn to the professionals... Who really chip in MKUL4 not to ask questions or as Klot said, the devil is programmed ... That's exactly what I need.
The advisor should be guaranteed to open and close trades in accordance with the logic. For a knowledgeable person this is a three-hour job in realtime mode... ... we will agree and do it. There it is written well, really not difficult and fast, just need a RELIABLE execution with handling all server errors, with pauses, etc. This will be discussed in the process of..... In turn, I start this EA on VPS and signals copying will be organized. Accordingly, I can connect to the project..... Maxim, tocxic, forexman77
I always say to my customers, don't make them copy my signals.
When I told my clients, I always say there are no fast EAs, it's just a painstaking job:
1 you need to formalize the TS and create the terms of reference
2 translate it into code understandable to the machine
3 check for errors in the execution of logic
4 check for execution errors in trading
5 to make additional changes according to the identified errors in logic and execution
6 to test it in the tester and on the real data
7 to add functions to the Expert Advisor, which it needs to work in the real trade
etc.
The programmer's function is only (2,5) to translate your terms of reference into the language of the machine and, if you find any inconsistencies in its work with your terms of reference, correct the errors; this is the end of the programmer's work, your Expert Advisor is working according to your terms of reference.
All the rest is your work on the creation, testing and improvement (upgrading) of your EA.
Best regards.
P.S. it may take you up to half a year to create a normal profitable Expert Advisor, or even more if you really need it.
P.S. it may take you up to half a year to create a normal profitable Expert Advisor, or even more if you really need it.
That's for sure.
And as the design at each stage requires adding new functionality that could not have been foreseen at the TOR stage or at earlier stages. All in all, it turns out to be much more complicated than expected. This is especially true for new strategies.
Exactly.
And as the design progresses, at each stage the need to add new functionality emerges, which could not have been envisioned at the TOR stage or at the previous stages. All in all, it turns out to be much more complicated than expected. Especially for new strategies.
Sincerely.