遗传算法优化初始值的BP神经网络股价预测程序，是智能优化算法的大作业供大家参考，结合了各种股价指标

clc close all C=xlsread('1.xlsx'); inputnum=10; hiddennum=4; outputnum=1; % Opening stock value for the day Open = C(1:750,4); Open = Open'; % 昨天的收盘价 shoupan = C(1:750,5); shoupan =shoupan'; % Highest stock value for the day High = C(1:750,2); High = High'; % Lowest stock value for the day Low = C(1:750,3); Low = Low'; % Closing stock value for the day Close = C(1:750,1); Close = Close'; %均线 ma5 = C(1:750,8); ma5 = ma5'; ma10 = C(1:750,9); ma10 = ma10'; ma20 = C(1:750,10); ma20 = ma20'; ma50 = C(1:750,11); ma50 = ma50'; %换手率 change=C(1:750,6); change = change'; %kdj d d=C(1:750,14); d = d'; %macd macd= C(1:750,19); macd=macd'; %% 遗传算法参数初始化 maxgen=10; %进化代数，即迭代次数 sizepop=10; %种群规模 pcross=[0.2]; %交叉概率选择，0和1之间 pmutation=[0.1]; %变异概率选择，0和1之间 %节点总数 numsum=10*4+4+4*1+1; lenchrom=ones(1,numsum); %个体长度 bound=[-3*ones(numsum,1) 3*ones(numsum,1)]; %个体范围 % Input vector of the input variables Input = {Open;shoupan;High; Low;ma5;ma10;ma20;ma50;change;macd}; Input = cell2mat(Input); % Construction of feed-forward neural network net = newff([minmax(Open);minmax(shoupan);minmax(High); minmax(Low);minmax(ma5);minmax(ma10);minmax(ma20);minmax(ma50);minmax(change);minmax(macd)], [abs(floor(4)), 1],{'purelin', 'purelin', 'transIm'},'traingdx'); % eveluating the performance of the neural network - using mse as % the measuring standard individuals=struct('fitness',zeros(1,sizepop), 'chrom',[]); %将种群信息定义为一个结构体 avgfitness=[]; %每一代种群的平均适应度 bestfitness=[]; %每一代种群的最佳适应度 bestchrom=[]; %适应度最好的染色体 %计算个体适应度值 for i=1:sizepop %随机产生一个种群 individuals.chrom(i,:)=Code(lenchrom,bound); %编码（binary和grey的编码结果为一个实数，float的编码结果为一个实数向量） x=individuals.chrom(i,:); %计算适应度 individuals.fitness(i)=fun(x,10,4,1,net,Input,Close); %染色体的适应度 end FitRecord=[]; %找最好的染色体 [bestfitness ,bestindex]=min(individuals.fitness); bestchrom=individuals.chrom(bestindex,:); %最好的染色体 avgfitness=sum(individuals.fitness)/sizepop; %染色体的平均适应度 %记录每一代进化中最好的适应度和平均适应度 trace=[avgfitness bestfitness]; %% 迭代求解最佳初始阀值和权值 % 进化开始 for i=1:maxgen % 选择 individuals=Select(individuals,sizepop); avgfitness=sum(individuals.fitness)/sizepop; %交叉 individuals.chrom=Cross(pcross,lenchrom,individuals.chrom,sizepop,bound); % 变异 individuals.chrom=Mutation(pmutation,lenchrom,individuals.chrom,sizepop,i,maxgen,bound); % 计算适应度 for j=1:sizepop x=individuals.chrom(j,:); %个体 individuals.fitness(j)=fun(x,10,4,1,net,Input,Close); end %找到最小和最大适应度的染色体及它们在种群中的位置 [newbestfitness,newbestindex]=min(individuals.fitness); [worestfitness,worestindex]=max(individuals.fitness); %最优个体更新 if bestfitness>newbestfitness bestfitness=newbestfitness; bestchrom=individuals.chrom(newbestindex,:); end individuals.chrom(worestindex,:)=bestchrom; individuals.fitness(worestindex)=bestfitness; %记录每一代进化中最好的适应度和平均适应度 avgfitness=sum(individuals.fitness)/sizepop; trace=[trace;avgfitness bestfitness]; FitRecord=[FitRecord;individuals.fitness]; end %% 把最优初始阀值权值赋予网络预测 % %用遗传算法优化的BP网络进行值预测 w1=x(1:inputnum*hiddennum); B1=x(inputnum*hiddennum+1:inputnum*hiddennum+hiddennum); w2=x(inputnum*hiddennum+hiddennum+1:inputnum*hiddennum+hiddennum+hiddennum*outputnum); B2=x(inputnum*hiddennum+hiddennum+hiddennum*outputnum+1:inputnum*hiddennum+hiddennum+hiddennum*outputnum+outputnum); net.iw{1,1}=reshape(w1,hiddennum,inputnum); net.lw{2,1}=reshape(w2,outputnum,hiddennum); net.b{1}=reshape(B1,hiddennum,1); net.b{2}=B2; % Maximum number of iterations net.trainparam.epochs = 65000; % Desired Tolerance value net.trainparam.goal = 1e-5; % learning rate initialisation net.trainparam.lr = 0.001; % using full data to train the neural network net.divideFcn ='dividetrain'; net = train(net, Input, Close); t = net(Input); perf = perform(net, Close, t); view(net); %% Testing the constructed neural network % Opening sample test data C_t = xlsread('1.xlsx'); Open_t = C_t(751:1500,4); Open_t = Open_t'; High_t = C_t(751:1500,2); High_t = High_t'; Low_t = C_t(751:1500,3); Low_t = Low_t'; Close_t = C_t(750:1500,1); Close_t = Close_t'; shoupan_t = C(751:1500,5); shoupan_t =shoupan_t'; ma5_t = C(751:1500,8); ma5_t = ma5_t'; ma10_t = C(751:1500,9); ma10_t = ma10_t'; ma20_t = C(751:1500,10); ma20_t = ma20_t'; ma50_t = C(751:1500,11); ma50_t = ma50_t'; %换手率 change_t=C(751:1500,6); change_t = change_t'; %d d_t= C(751:1500,14); d_t=d_t'; %macd macd_t= C(751:1500,19); macd_t=macd_t'; Input_t = {Open_t; shoupan_t;High_t; Low_t;ma5_t;ma10_t;ma20_t;ma50_t;change_t;macd_t}; Input_t = cell2mat(Input_t); answer = ones(1,750); answer_t = ones(1,750); for i=1:750 answer(i) = net([Input_t(1, i); Input_t(2,i); Input_t(3, i); Input_t(4, i);Input_t(5, i);Input_t(6, i);Input_t(7, i);Input_t(8, i);Input_t(9, i);Input_t(10, i)]); answer_t(i) = Close_t(i); end x = 1:750; plot(x, answer,'r' ,x, answer_t(1:750),'b'); legend('真值','预测值') xlabel('数据点'); ylabel('股票价格'); title('神经网络预测股价'); acc=corr(answer(1:end)',answer_t(1:end)'); pr=sprintf('Acc is %2f',acc); msgbox(pr) figure(2) [r c]=size(trace); plot([1:r]',trace(:,1),'b',[1:r]',trace(:,2),'k'); title(['适应度曲线 ' '终止代数＝' num2str(maxgen)]); xlabel('进化代数');ylabel('适应度'); legend('平均适应度','最佳适应度');