价格预测基于matlab SVM价格预测【含Matlab源码 180期】.zip

clear; clc; %load financial data of the stock price of Apple company %The data is from Nov 18 1982-Nov 18 2012 %The data contains six collums:Open, High, Low, Close, Volume, Adj Close sh = dlmread('yahoo.csv'); %The data needs to flip because the data is from latest to earliest. sh = flipdim(sh,1); %extract data [m,n] = size(sh); ts = sh(2:m,1); tsx = sh(1:m-1,:); original = ts(length(sh)*0.7+1:end,:); % Draw the original graphic of the stock price figure; plot(ts,'LineWidth',1); title('Yahoo Stock Price(1996.4.12-2012.11.16) before mapping','FontSize',12); grid on; fprintf('Plot the stock price before mapping.\n'); fprintf('Program paused. Press enter to continue.\n'); pause; %data preprocessing ts = ts'; tsx = tsx'; % mapminmax is an mapping function in matlab %Use mapminmax to do mapping [TS,TSps] = mapminmax(ts); % The scale of the data from 1 to 2 TSps.ymin = 1; TSps.ymax = 2; %normalization [TS,TSps] = mapminmax(ts,TSps); % plot the graphic of the stock price after mapping figure; plot(TS,'LineWidth',1); title('Yahoo Stock price after mapping','FontSize',12); grid on; fprintf('\nPlot the stock price after mapping.\n'); fprintf('Program paused. Press enter to continue.\n'); pause; % Transpose the data in order to meet the requirement of libsvm fprintf('\n Initializing.......\n'); TS = TS'; [TSX,TSXps] = mapminmax(tsx); TSXps.ymin = 1; TSXps.ymax = 2; [TSX,TSXps] = mapminmax(tsx,TSXps); TSX = TSX'; %split the data into training and testing n1 = length(TS)*0.7; train_label = TS(1:n1,:); train_data = TSX(1:n1,:); test_label = TS(n1+1:end,:); test_data = TSX(n1+1:end,:); fprintf('Begin the two round regressions to tune the parameter.\n'); fprintf('Program paused. Press enter to continue.\n'); pause; % Find the optimize value of c,g paramter % Approximately choose the parameters: % the scale of c is 2^(-5),2^(-4),...,2^(10) % the scale of g is 2^(-5),2^(-4),...,2^(5) [bestmse,bestc,bestg] = svmregress(train_label,train_label,-5,10,-5,5,3,1,1,0.0005); % Display the approximate result disp('Display the approximate result'); str = sprintf( 'Best Cross Validation MSE = %g Best c = %g Best g = %g',bestmse,bestc,bestg); disp(str); fprintf('\nFinish the first round tuning and begin the final round regression and print the final parameters.\n'); fprintf('Program paused. Press enter to continue.\n'); pause; % Choose more precise parameter according to the graphic of previous step: % the scale of c is 2^(0),2^(0.3),...,2^(10) % the scale of g is 2^(-2),2^(-1.7),...,2^(3) [bestmse,bestc,bestg] = svmregress(train_label,train_data,0,10,-2,3,3,0.3,0.3,0.0002); disp('Display the final parameter result'); str = sprintf( 'Best Cross Validation MSE = %g Best c = %g Best g = %g',bestmse,bestc,bestg); disp(str); fprintf('\nProgram paused. Press enter to continue.\n'); fprintf('Predict the stock price of the training data and compare to the grand truth.\n'); pause; %Do training by using svmtrain of libsvm cmd = ['-c ', num2str(bestc), ' -g ', num2str(bestg) , ' -s 3 -p 0.01']; model = libsvmtrain(train_label,train_data,cmd); %Do predicting by using svmpredict of libsvm predict= libsvmpredict(test_label,test_data,model); predict = mapminmax('reverse',predict,TSps); % Display the result of SVM Regression str = sprintf( 'MSE = %g R = %g%%',mse(2),mse(3)*100); disp(str); figure; hold on; plot(original,'LineWidth',1); plot(predict,'r','LineWidth',1); legend('Original Price','Predict Price','FontSize',10); hold off; grid on; snapnow;