"NLD_BackProp.mqh" contains the implementation of a very simple Artificial Neural Network that has one hidden layer and uses back-propagation for training/learning. The advantage of this implementation is that it is very short, which makes it easy for anyone to get started with ANNs.
"NLD_BPChart.mq5" is a custom price chart, painting price movements as a red/green line and using the Artificial Neural Network (implemented in "NLD_BackProp.mqh") to try predicting where the Close price might be going next, relative to the last Open price. It projects the next X Close prices (default = 10) into the future by using the last Y Open prices (default = 200) from the past, optimizing network weights with every new (completed) candle received.
I wouldn't call the output of the Artificial Neural Network a "prediction", but simply an "opinion", which I would NOT take seriously when trading, so do NOT get too excited about it. But ... if you find Artificial Neural Networks interesting, it might be fun to watch how an Artificial Neural Network tries to optimize its outputs as new data arrives. There are many ways to use Artificial Neural Networks and this is just a very simple example, so have fun experimenting!
Both files are rather simple and do NOT have any external dependencies on other libraries or files.
USE AT YOUR OWN RISK!
