Discussion of article "Forecasting with ARIMA models in MQL5"
Hello Francis!
An excellent article! As I am converting an MQ4 EA that use several averages, I am going to study your code to determine the best integration and will be back in touch later
CapeCoddah
I did everything I could to get no result.
The article has attached a screenshot from a tester that has a wild profit factor but a low recovery factor. What it can be and how the chart looks like is a secret.
If the author managed to get something, it must be a special secret.
By the way, the settings differ from what is in the article. All in all, strange content.
Thanks for your very informative article. Can I predict more than one step ahead?
I guess just increasing "NumberOfPredictions" isn't enough, is it?
@ndnz2018: habe vor kurzem folgende Klasse für ARIMA-Modell heruntergeladen: https://www.mql5.com/en/articles/12798
I was really irritated when I saw that the bot showed hit rates of over 90% in several time frames for simple AR models (20,1,0). I also found other articles online where ARIMA models were presented with accuracies of over 90%. Yet you can read in any book on financial mathematics that increases (returns) show no significant auto-correlation. I have also calculated the auto-correlation Corr(r(t),r(t-d)) for different lags myself and it is true that there is no correlation. How can that be? Is auto-correlation perhaps defined differently in the case of ARIMA models? Actually, I always thought that autoregression is a simple regression on the previous values. Am I perhaps seeing this too simply ?
I hope someone can clear my head.
Many thanks in advance
ndnz
There is an EA in the article. Why don't you run it on a demo account and see if the values are confirmed.
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
You agree to website policy and terms of use
New article Forecasting with ARIMA models in MQL5 has been published:
In this article we continue the development of the CArima class for building ARIMA models by adding intuitive methods that enable forecasting.
It is well known that ARIMA models rely on temporal dependences in a dataset. Therefore to make one or more predictions we need to feed the model a series of input data. The specification of the model determines the minimum size of the input series. Knowning this, it becomes obvious that if the input series is inadequate it will not be possible to make any predictions or at the very least the predictions will not be reflective of the applied model. Different types of ARIMA models make varying demands on the size of the input series beyond just the order of the model.
Implementing predictions for pure autoregressive models is trivial as all that is required are inputs equal to the largest lag of the model. Mixed models that use moving average terms create problems when making forecasts. We have no actual error or innovation series yet. To overcome this we must first decide how the initial values of errors will be calculated.
This process involves first using any available model parameters to get the initial state of the model which excludes any of the moving average terms, as they are assumed to be 0 at this stage. Then the known series values are used to calculate initial error values by cycling through an number of redundant predictions.These initial predictions are redundant because they will have nothing to do with the final prediction(s) we are ultimately interested in. This obviously puts more demands on the number of inputs needed for prediction. The critical thing to appreciate here is how many redundant prediction cycles should be peformed in order to come up with suitable error series values to make valid predictions.
Author: Francis Dube