Modeling and Computing of Stock Index Forecasting Based on Neural Network and Markov Chain

Algorithm 1

Part of the code of MATLAB.

sample = xlsread(‘data.xls’)

[m,n] = size(sample);

ts = sample(2:m,4);

tsx = sample(1:m-1,:);

TS = ts’;

TSX = tsx’;

[Pn1,minp1,maxp1] = premnmx(TS)

[Pn2,minp2,maxp2] = premnmx(TSX)

rand(‘state’,0);

PR = [−1,1; −1,1; −1,1; −1,1; −1,1];

net=newff(PR,[11,1],{‘tansig’,‘purelin’},‘traingdx’,‘learngdm’,‘mse’)

net.trainparam.show = 50;

net.trainparam.lr = 0.1;

net.trainparam.mc = 0.9;

net.trainparam.epochs = 10000;

net.trainparam.goal = 0.005;

net = init(net);

net = train(net,Pn2(:,1:41),Pn1(1:41));

% Forecast

test = Pn2(:,42:end);

Y = sim(net, test)

P = postmnmx(Y, minp1, maxp1)

e = (sample(42:m-1,4)’-P)./sample(42:m-1)

% Calculate total error of network

res = norm(e)

figure(1)

t = 42: m-1

% Draw figure

plot((1:m-1), sample(1:m-1),‘-*’,t, P, ‘o’)