cm_mod1_new.m

Neural Network in Finance (神经网络在金融界:赢得预言性的优势)全部原码。内容包括预测与估计

function [error, yhat, pderiv,neuron1] = cm_mod1(beta);
global data1 squasher maxx minx maxy miny malags neuronxarg_inf cthres_inf;
y = data1(:,1);
x = data1(:,2:end);
[nx, cx] = size(x);
% beta(1:2*cx+3) = abs(beta(1:2*cx+3));
if squasher == 1,
    yy = (y - miny)/(maxy-miny);
    for i = 1:cx, xx(:,i) = (x(:,i)-minx(i))/(maxx(i)-minx(i));
    end
else yy = y; xx = x;
end
if squasher == 1,
    cxarg_inf = (cthres_inf-minx(neuronxarg)) / (maxx(neuronxarg_inf)-minx(neuronxarg_inf));
else cxarg_inf = cthres_inf;
end

ny = length(yy);
yhat1 = yy; 
xx1 = xx * abs(beta(1:cx))' + ones(ny,1) * beta(cx+1);
ehat(1:malags,1) = zeros(malags,1); 
neuron3(1:malags,1) = zeros(malags,1); 
for i = malags+1:ny, 
neuronx = abs(beta(cx+2))* (xx(i-1,neuronxarg_inf)-cxarg_inf);
neuron1(i,:) =  2 ./ (1 + exp(-2 * neuronx)) -1;
EXX = ehat(i-malags:i-1,:);
yhat1(i,:) = xx1(i,:)  + neuron1(i,:) * abs(beta(cx+3)) * xx1(i,neuronxarg_inf) + beta(cx + 4:cx+malags+3) * EXX;
ehat(i,:) = yy(i,:) - yhat1(i,:);
end;
nparm =  cx + malags+3;
error = yy - yhat1;
error = mean(error .^2);
sigma = error/ nparm;  
T = length(yhat1);          
loglik = -.5 * T * log(2 * pi) - .5 * T * log(sigma) - .5 * error /sigma;
error = -loglik;
if squasher == 1,
    yhat = yhat1 * (maxy-miny) + miny;
else yhat = yhat1;
end;
for i = 1:cx,
for j = 2:ny,
    pderiv(j,i) =   abs(beta(i));
end;
end;
for j = 2:ny,
    pderiv(j,neuronxarg_inf) = abs(beta(neuronxarg_inf)) + neuron1(j) * abs(beta(cx+3)); 
end;