timeseries forecasting for 10 years ahead ANN

Hi,

i would like to forecast the Cl parameter for 10 years ahead. But keep getting error : Index in position 2 exceeds array bounds. Index must not exceed 1. by the way , when running the code, i have remove the heading, words from the excel. but still getting the error

Error in preparets (line 317)

xi = xx(:,FBS+((1-net.numInputDelays):0));

Error in timee (line 25)

[X,inputStates,layerStates,T] = preparets(net,inputSeries,{},targetSeries);

Please help to check which part that i did wrong.

data = readmatrix ("104CPr.csv");

Input = data(:,1:(end-1));

Target = data(:,end);

X = con2seq(Input)';

T = con2seq(Target)';

%% Data Preparation

N = 10; % Multi-step ahead prediction

% Input and target series are divided in two groups of data:

% 1st group: used to train the network

inputSeries = X(1:end-N);

targetSeries = T(1:end-N);

%% Network architecture

delay = 2;

inputDelays = 1:2;

feedbackDelays = 1:2;

hiddenLayerSize = 10;

trainFcn = 'trainbr';

% Network Creation

net = narxnet(inputDelays,feedbackDelays,hiddenLayerSize,'open', trainFcn);

[X,inputStates,layerStates,T] = preparets(net,inputSeries,{},targetSeries);

net.divideParam.trainRatio = 70/100;

net.divideParam.valRatio = 15/100;

net.divideParam.testRatio = 15/100;

[net,Y] = train(net,X,T,inputStates,layerStates);

outputs = net(X,inputStates,layerStates);

errors = gsubtract(T,Y);

performance = perform(net,T,Y);

view(net)

delay = 2;

netc = closeloop(net);

netc.name = [net.name ' - Closed Loop'];

view(netc)

[xc,xic,aic,tc] = preparets(netc,inputSeries,{},targetSeries);

yc = netc(xc,xic,aic);

closedLoopPerformance = perform(netc,tc,yc);

% Early Prediction Network

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);

%%5. Multi-step ahead prediction

inputSeriesPred = [inputSeries(end-delay+1:end),inputSeriesVal];

targetSeriesPred = [targetSeries(end-delay+1:end), con2seq(nan(1,N))];

[Xs,Xi,Ai,Ts] = preparets(netc,inputSeriesPred,{},targetSeriesPred);

yPred = netc(Xs,Xi,Ai);

perf = perform(net,yPred,targetSeriesVal);

figure;

plot([cell2mat(targetSeries),nan(1,N);

nan(1,length(targetSeries)),cell2mat(yPred);

nan(1,length(targetSeries)),cell2mat(targetSeriesVal)]')

legend('Original Targets','Network Predictions','Expected Outputs');

Answer 1

[Sandeep Mishra]

Hi Nur,

Upon debugging the code snippet, I realised that the variable 'X' is a 7*1 cell array while the variable 'T' is 1*1 cell array.

The root cause of the issue arises due to the variable ‘N’ (N=10) used for computing 'inputSeries' and 'targetSeries', which exceeds the sizes of variable 'X' and 'T', resulting in empty arrays (0x1 and 1x0).

To resolve the issue, you can adjust the value of ‘N’ so that 'inputSeries' and 'targetSeries' are appropriately sized.

For a detailed example on using the 'preparets' function, you can refer to the following MathWorks documentation: https://www.mathworks.com/help/releases/R2024b/deeplearning/ref/preparets.html#f8-3421485

I hope this helps!

[Nur Farahin Che Nordin]

Hi Sandeep,

I have adjust the value of N, however still getting the same error. Would you mind to help on this

[Sandeep Mishra]

Hi Nur, What value of N are you using now?

[Nur Farahin Che Nordin]

i have change to 2, still getting the same error.

then, i tried to use this code instead of previous code

N = 10; % Multi-step ahead prediction

% Input and target series are divided in two groups of data:

% 1st group: used to train the network

i = N+1: length (data)

inputSeries = [X;data(i-N:i-1)];

targetSeries = [T;data(i)];

now, im getting the error Error using preparets

Inputs{1,8} and Inputs{1,1} have different numbers of rows.

Error in timee (line 26)

[X,inputStates,layerStates,T] = preparets(net,inputSeries,{},targetSeries);

please help, im drained and lost right now

[Nur Farahin Che Nordin]

i tried changing to simple code as below but still not working, the code stucked half way with error Error using network/sim

Input data sizes do not match net.inputs{1}.size.

Error in indexing (line 15)

otherwise, v = sim(vin,subs{:});

Error in timeseriess (line 24)

forecast(i) = net(lastValue);

data = readmatrix ("104CPr.csv");

% Define the number of years to forecast

forecastYears = 10;

% Prepare the data for training

X = data(1:end-1); % Inputs

T = data(2:end); % Targets

% Create and configure the neural network

net = feedforwardnet(10); % 10 hidden neurons

net = configure(net, X', T');

% Train the neural network

[net, tr] = train(net, X', T');

% Forecast the next 10 years

forecast = zeros(1, forecastYears);

lastValue = data(end);

for i = 1:forecastYears

forecast(i) = net(lastValue);

lastValue = forecast(i);

end