Machine learning in trading: theory, models, practice and algo-trading - page 3565
![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
programmed a simple variant. some primitive simulation of the author's idea.
Arima starts a forecast in a sliding window with a piece of random series (first 100 points) and adds a new forecast point + noise at the end and the window shifts by +1. Eventually the series consists of only forecasts and a new forecast is made exclusively from the series of old forecasts.
We have got quite plausible series
And sometimes the model closes on the trend and we get a certain sticking in the forecast (strong unanticipated trend).
Which is also peculiar to the market.
And sometimes the model closes on the trend and we get a certain sticking in the forecast (a strong trend that is not expected).
Which is also peculiar to the market
Here is another cool example of how the model started to accelerate itself.
Just like bitcoin ))
And sometimes the model closes on the trend and we get a certain sticking in the forecast (a strong trend that is not expected).
Which is also peculiar to the market
Sometimes it seems that the model is stuck like on the picture.
Well, as if it is a flat and that's it, the model draws a flat and predicts a flat further only a flat...
But no....
man, it turned out to be quite interesting.)
as they say, "instead of a thousand words.
"instead of a thousand, a plum." :-)
"instead of a thousand, it's a plum" :-)
Well, the likes are finally coming in handy
Sometimes it looks like the model is stuck like in the picture.
Well, as if it were a flat and that's it, the model draws a flat and predicts a flat further only a flat...
But no....
man, it turned out to be quite interesting.)
The Elliott Wave has turned out again. If the first correction was long and shallow, the second correction is usually shorter and deeper :)
An Elyot wave is just two sine waves with the right characteristics... There's no magic in it like the other patterns.
I have created an algorithm for forecasting on the hourly and even on the minutes....
It is actually possible to forecast 24 points a day ahead.