NNET is an R package that helps you make a single hidden layer feed forward perceptron Neural Network model.
You select the inputs that you want to have.
NNET is fast in calculating the model.
But if you select too many inputs, you will get an error that says too many weights need to be calculated.
I have shown in the blog post how you will use this NNET R package to develop a feed forward neural network to model daily and weekly candle.
You can use this model to predict the high, low and close of the weekly and daily candle on Sunday.
So next week you have a rough idea how much the price is going to move.
This will help you in opening trades in the right direction.