Neural network traiding system

I'd like to create a neural network to forecast the trend of the market ( up or down respectively 1 and 0);

I have a database of historical data provided by a broker and I want to understand 2 things:

First, if I use the pre-programmed neural network using 'nnstart' it's sufficient or i have to create the neural network by myself?

Second, do I have to use the narx function or which one?

Answer 1

[Srivardhan Gadila]

You can start working on your problem by using nnstart. Set aside some samples of your historical data for validating the trained network and based on the validation results you can decide on creating a nerual network by yourself.

If your goal is to predict series y(t) given d past values of y(t) and another series x(t) then go for NARX or if your goal is to predict series y(t) given d past values of y(t) then go for NAR else your goal is to predict series y(t) given d past values of series x(t) then go for Nonlinear Input-Ouput.

The following link might help you in the process: Neural Network Time-Series Prediction and Modeling