CODEWITHSUNDEEP
matlabneural-networktime-series
James
James

Reputation: 31738

How can MATLAB's narnet be used to predict future values of a variable

Given a set of past values of a variable, how can future ones be predicted with MATLAB's narnet?

An example given my MATLAB's Neural Net Time Series App is as follows:

T = oil_dataset; 
net = narnet(1:2,10); 
[Xs,Xi,Ai,Ts] = preparets(net,{},{},T); 
net = train(net,X,T,Xi,Ai); 
view(net) Y = net(Xs,Xi,Ai) 
plotresponse(T,Y)

How could say the next 10 values be predicted of the oil_dataset?

Upvotes: 0

Views: 1403

Answers (1)

Eskapp
Eskapp

Reputation: 3635

If you use the Matlab NSTTool, at the very last step, you can automatically generate a script with examples (click on "Advanced script" box). In this code, there is an example of how to predict multiple values. The function used is cnet.

This is the part of the code I generated for my own net:

    [x1,xio,aio,t] = preparets(net,{},{},T);
    [y1,xfo,afo] = net(x1,xio,aio);
    [netc,xic,aic] = closeloop(net,xfo,afo);
    [y2,xfc,afc] = netc(cell(0,20),xic,aic); % Predict next 20 values

Upvotes: 1

Related Questions