MATLAB: NARX – Next Day Prediction of Stock Prices

narx netneural networkstime delay

Hi,

I am using a Time Dalay NARX Neural Network to predict the next day prices of stocks from a particular industry sector (marine and offshore, Singapore Exchange). I am using the attached dataset along with the following code for the prediction attempt. The prediction results are quite bad.

I had read that autocorelation and cross-correlation is required to get the number of delays, but I have been unsuccessful in implementing it. Could you modify the code below to implement auto-correlation and cross-correlation for getting the appropriate delays? It would be great if you have any suggestions to improve prediction accuracy. (Using additional data, different training algorithm, using additional algorithms etc.)

    data=csvread('keppelcorp.csv',1,1);
    inputSeries = tonndata(data(2:end,:),false,false);
    targetSeries = tonndata(data(1,:),false,false);
    input = data(1:end,2:end);
    target = data(1:end,1);
    N=int32(0.7*size(input,1));  % how much data to use for training
    inputSeries = tonndata(input(1:N,:),false,false);
    targetSeries = tonndata(target(1:N,:),false,false);
    inputSeriesVal = tonndata(input((N+1):end,:),false,false);
    targetSeriesVal = tonndata(target((N+1):end,:),false,false);
    inputDelays = 1:10;
    feedbackDelays = 1:10;
    hiddenLayerSize = 20;
    net = narxnet(inputDelays,feedbackDelays,hiddenLayerSize);
    [inputs,inputStates,layerStates,targets] = preparets(net,inputSeries,{},targetSeries);
    net.divideFcn = 'divideblock';  % Divide data in blocks
    net.divideMode = 'time';  % Divide up every value
    net.trainFcn = 'trainlm';  % Levenberg-Marquardt
    % Train the Network
    [net,tr] = train(net,inputs,targets,inputStates,layerStates);
    plotperform(tr)
    Y = net(inputs,inputStates,layerStates); 
     % Prediction Attempt
    N=size(inputSeriesVal,2);
    delay=length(inputDelays); 
    inputSeriesPred  = [inputSeries(end-delay+1:end),inputSeriesVal];
    targetSeriesPred = [targetSeries(end-delay+1:end), con2seq(nan(1,N))];
    netc = closeloop(net);
    [Xs,Xi,Ai,Ts] = preparets(netc,inputSeriesPred,{},targetSeriesPred);
    yPred = netc(Xs,Xi,Ai);
    perf = perform(net,yPred,targetSeriesVal);
    perf_mae = mae(net,targetSeriesVal, yPred);
    perf_mse = mse(net,targetSeriesVal, yPred);
    close all;
    targetSeriesVal = targetSeriesVal(2:end);
    targetSeriesVal(end+1) = targetSeriesVal(end);
    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')
    figure;
    plot([cell2mat(yPred);cell2mat(targetSeriesVal)]')
    legend('Network Predictions','Expected Outputs')
    disp('Performance is')
    perf
    disp('Mean absolute error performance is')
    perf_mae
    disp('Normmalized MSE Performance is')
    perf_mse
    disp('Next day closing price was forecasted to')
    yPred(end)
    csvwrite('Real 1st column and 2nd column Predicted Prices.csv',[targetSeriesVal',yPred']);

Best Answer

> I am using a Time Dalay NARX Neural Network to predict the next day prices of stocks from a

> particular industry sector (marine and offshore, Singapore Exchange). I am using the attached

> dataset along with the% following code for the prediction attempt. The prediction results are

> quite bad. I had read that autocorelation and cross-correlation is required to get the number

> of delays, but I have been unsuccessful in implementing it. Could you modify the code below

> to implement auto-correlation and cross-correlation for getting the appropriate delays?

I'm surprised to hear that. If you just search in the NEWSGROUP and ANSWERS using

greg NNCORR

NEWSGROUP 11 hits

ANSWERS 53 hits

> It would be great if you have any suggestions to improve prediction accuracy. (Using

> additional data, different training algorithm, using additional algorithms etc.)

My answer is simple:

Take a look at some of my posts.

Hope this helps.

Thank you for formally accepting my answer

Greg

Related Solutions

MATLAB: 20 Days ahead forecasts using NAR

Try

 FD = 1:20;
 net = narnet(FD,H=10);
 net.divideFcn = 'divideblock';  % Never use 'dividerand' for timeseries.

For details see

 help narnet
 doc narnet

and search the NEWSGROUP and ANSWERS using

 greg narnet

Hope this helps

Thank you for formally accepting my answer

Greg

MATLAB: Hi, i am using NARX todo multi step prediction of a daily stock market index (Sensex 2003×1 matrix) using another one as input (Nifty 2003×1 matrix). I am having problem with the close loop

%% 1. Importing data % Matrix of 2003x1 each are daily stock market indices data of Nifty & Sensex

 > load Nifty.dat;
 > load Sensex.dat;

% To scale the data it is converted to its log value:

 > lognifty = log(Nifty); 
 > logsensex = log(Sensex);
 > X = tonndata(lognifty,false,false); 
 > T =  tonndata(logsensex,false,false);

%% 2. Data preparation

 > N = 300; % 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); % 2nd group: this is the new data used for simulation. inputSeriesVal will % be used for predicting new targets. targetSeriesVal will be used for % network validation after prediction

Notation:

 data    = design + test
 design = training + validation

Val subsets are used repetetively with Trn subsets to DESIGN a net with a good set of training parameters (e.g., input delays, feedback delays, number of hidden nodes, stopping epoch, etc). The best of multiple designs is, typically, based on indirectly minimizing MSEval.

After the best design is chosen, the nondesign Test subset is used to estimate generalization performance on nondesign data.

By DEFAULT, the data will be divided RANDOMLY into THREE trn/val/tst subsets according to

 dividerand( 2003, 0.7, 0.15, 0.15 )

I disagree with the use of dividerand for uniformly spaced time-series. Replace with one of the other divide functions. (When Nval=Ntst =0, I use 'dividetrain'. Otherwise I use , 'divideblock' or 'divideind' to maintain uniform spacing);

 > inputSeriesVal  = X(end-N+1:end);
 > targetSeriesVal = T(end-N+1:end); % This is generally not available

Change "Val" to "Test" since the subsets are only used for performance evaluation (NOT "validation") and not design.

Since a NNTBX BUG will not allow a test subset without a validation subset and visa versa, there are two options

 1. Use trn/val/tst (Nval=Ntst = 300) and 'divideblock' or 'divideind'  
   (recommended) 
 2. a.Remove the tst subset (Ntst = 300) from training, 
    b. Do not use a val set (Nval=0) 
    c. Use 'dividetrain' to only train on training data(Ntrn = 1703). 
    d. Calculate the test subset performance separately

%% 3.

 > Network Architecture delay = 2; 
 > neuronsHiddenLayer = 50;

Use the autocorrelation function to determine the significant feedback delays. Use the crosscorrelation function to determine the significant input delays.

% Network Creation

 > net = narxnet(1:delay,1:delay,neuronsHiddenLayer);

%% 4. Training the network

 > [Xs,Xi,Ai,Ts] = preparets(net,inputSeries,{},targetSeries); 
% > net = train(net,Xs,Ts,Xi,Ai);
 [ net tr Ys Es Xf Af ] = train(net,Xs,Ts,Xi,Ai);
 tr = tr     % To obtain important info
 > view(net) 
 > Y = net(Xs,Xi,Ai); % Performance for the series-parallel implementation, only

% one-step-ahead prediction

 > perf = perform(net,Ts,Y);

%% 5. Multi-step ahead prediction

 >inputSeriesPred  =[inputSeries(end-delay+1:end),inputSeriesVal]; 
 >targetSeriesPred = [targetSeries(end-delay+1:end), con2seq(nan(1,N))]; 
 >netc = closeloop(net); 
 >view(netc)

Check netc on previous data. If performance is bad, improve it by training netc on the previous data.

 >[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')

% Network predictions are coming very bad.. I guess there is some problem % with the close loop's initial input states and initial layer states. % please help.

1. Optimize ID and FD

2. Use trn/val/trn with 'divideblock' or 'divideind'

3. Compare netc and net performance on openloop data

4. If necessary, use train on netc.

5. Then consider nondesign data

Hope this helps

Thank you for formally accepting my answer

Greg

Best Answer

Related Solutions

MATLAB: 20 Days ahead forecasts using NAR

MATLAB: Hi, i am using NARX todo multi step prediction of a daily stock market index (Sensex 2003×1 matrix) using another one as input (Nifty 2003×1 matrix). I am having problem with the close loop

Related Question