MATLAB: Prediction Problem using NARX neural network

narx prediction

I would like to conduct the multistep prediction by using NARX network. The network is trained. However, I am confused how to use it to predict the next values.

For example, output y(t+1) can be predicted by the trained network. However, in order to predict y(t+2), I need y(t+1) and u(t+1) (external input). But I did not know u(t+1) either. Should I predict u(t+1) first of all?

Thanks for answering

Best Answer

If you are using NARX, the predictions stop when the inputs stop.

If you want to predict further, change to NARNET that does not require an input.

Hope this helps.

Thank you for formally accepting my answer

Greg

Related Solutions

MATLAB: How to predict future responses y(t + 1) from the training of a narxnet network with past data of x (t) and y (t)? (NARXNET)

YOU DO NOT HAVE X and Y !!!

YOU HAVE X and T where

T = Ydesired

Hope this helps

Thank you for formally accepting my answer

Greg

MATLAB: Prediction using narx Network

The Fibonacci series does not result from an input/output relationship. It is autoregressive

Either y(1:2) = [ 0 1 ] or y(1:2) = [ 1 1 ] and then

 y(n+1) = y(n) + y(n-1)

Obviously this can be implemented with a NARNET WITH NO HIDDEN NODES.

Hope this helps.

Thank you for formally accepting my answer

Greg

Best Answer

Related Solutions

MATLAB: How to predict future responses y(t + 1) from the training of a narxnet network with past data of x (t) and y (t)? (NARXNET)

MATLAB: Prediction using narx Network

Related Question