how can i predict one week later(multistep prediction), this code just predicts one step in the future please make my code right(it's about wind speed prediction with 3 input parameter(pressure, humidity, temperature) and one target(wind speed)

load('input.mat'); Input_Parameter = tonndata(inputData(:,(1:3)),false,false);

Target_Parameter = tonndata(inputData(:,1),false,false);

inputSeries = Input_Parameter; targetSeries = Target_Parameter;

inputDelays = 1:4; feedbackDelays = 1:4; hiddenLayerSize = 10; net = narxnet(inputDelays,feedbackDelays,hiddenLayerSize);

% Prepare the Data for Training and Simulation % The function PREPARETS prepares time series data % for a particular network, shifting time by the minimum % amount to fill input states and layer states. % Using PREPARETS allows you to keep your original % time series data unchanged, while easily customizing it % for networks with differing numbers of delays, with % open loop or closed loop feedback modes. [inputs,inputStates,layerStates,targets] = preparets(net,inputSeries,{},targetSeries);

% Set up Division of Data for Training, Validation, Testing net.divideParam.trainRatio = 70/100; net.divideParam.valRatio = 15/100; net.divideParam.testRatio = 15/100;

% Train the Network [net,tr] = train(net,inputs,targets,inputStates,layerStates);

% Test the Network outputs = net(inputs,inputStates,layerStates); errors = gsubtract(targets,outputs); performance = perform(net,targets,outputs);

% View the Network view(net) %------------------------------------------------ outputs = cell2mat(outputs); N=length(outputs); figure(1), hold on plot( 1:N, outputs, 'LineWidth', 2) plot( 1:N, outputs, 'ro', 'LineWidth', 2) legend( ' TARGET ', ' OUTPUT ' ) title( ' NARXNET EXAMPLE ' ) %--------------------------------------------------

netc = closeloop(net); netc.name = [net.name ' - Closed Loop']; view(netc) [xc,xic,aic,tc] = preparets(netc,inputSeries,{},targetSeries); yc = netc(xc,xic,aic); perfc = perform(netc,tc,yc);

nets = removedelay(net); nets.name = [net.name ' - Predict One Step Ahead']; view(nets) [xs,xis,ais,ts] = preparets(nets,inputSeries,{},targetSeries); ys = nets(xs,xis,ais); earlyPredictPerformance = perform(nets,ts,ys);

ys = cell2mat(ys); M=length(ys); figure(1), hold on plot( 1:M, ys, 'LineWidth', 1) plot( 1:M, ys, 'ro', 'LineWidth', 1) legend( ' TARGET ', ' OUTPUT ' ) title( ' NARXNET EXAMPLE ' )

Answer 1

[Greg Heath]

Insuficient info;

HOW LONG IS ONE DELAY ...

1 second? 1 minute? 1 hour? 1 day?...

although you can recursively predict multiple delays ahead, errors will accumulate and the predictions become meaningless.

There is no magic formula that specifies how many multiples of the maximum delay can be used for predictions before errors accumulate to overwhelm accuracy.

Hope this helps

Thank you for formally accepting my answer

Greg