mql5

The article Implementing an ARIMA training algorithm in MQL5, describes the CArima class for building ARIMA models. Although it is technically possible to use the class as it is to apply a model and make predictions, it is not intuitive. In this article we will address this shortcomming and extend the class to enable easier to use methods for applying models to make predictions. We will discuss some of the complications related to implementing predictions as well as some new features added to the class. To conclude we will use the complete class to build a model , use it to predict forex prices by applying it to an expert advisor and indicator.

more...