Original question from Azhahra Vindy Ariesta retrieved from Google Cache:
How can I change the variable x0 to x2,x3,x4,...xn . So it still produces the same pattern?
Hi guys, your help will mean a lot.
So I'm making prediction modeling with Gray GM(1,N) method. In the formula it is necessary to define the values of the variables X and Y. Where in my program, the variable Y is A and the variable X is x0. The X variable that I have there are 7 factors that affect the Y variable.
But what my program is asking for, each X variable should be separated. So in this case, I'm confused about changing the loop function and changing what it should have been
x0 = [408, 454, 941, 17162, 35687, 19932, 61510, 81608, 122599, 199153, 156101, 120015;
124171, 190032, 219189, 152435, 130347, 191144, 133795, 53560, 31457, 26162, 15893, 20052;
70409, 207282, 1494766, 1433620, 1110495, 2224241, 2359721, 2913705, 2800503, 912552, 600338, 297624;
204975, 190683, 122735, 60850, 46974, 33007, 65187, 79223, 53214, 48594, 26030, 29108;
67018, 38201, 19335, 11657, 12855, 6542, 6611, 4101, 2953, 2723, 1313, 1393;
190988, 182134, 120311, 60621, 44094, 31392, 53312, 71792, 50007, 44888, 24092, 27284;
72365, 46900, 21759, 11886, 15735, 8157, 18336, 11826, 6160, 5761, 3251, 3537];
then changed to:
x0 = [408, 454, 941, 17162, 35687, 19932, 61510, 81608, 122599, 199153, 156101, 120015];
x1 = [124171, 190032, 219189, 152435, 130347, 191144, 133795, 53560, 31457, 26162, 15893, 20052];
x2 = [ 70409, 207282, 1494766, 1433620, 1110495, 2224241, 2359721, 2913705, 2800503, 912552, 600338, 297624];
etc. to xn.
This is my current program
clear
clc
A =[85983, 70017, 42320, 26565, 21439, 113409, 271185, 35930, 7182, 3775, 2407, 1350]; %(X0(1))
x0 =[408, 454, 941, 17162, 35687, 19932, 61510, 81608, 122599, 199153, 156101, 120015; %(X0(2))
124171, 190032, 219189, 152435, 130347, 191144, 133795, 53560, 31457, 26162, 15893, 20052; %(X0(3))
70409, 207282, 1494766, 1433620, 1110495, 2224241, 2359721, 2913705, 2800503, 912552, 600338, 297624; %(X0(4))
204975, 190683, 122735, 60850, 46974, 33007, 65187, 79223, 53214, 48594, 26030, 29108; %(X0(5))
67018, 38201, 19335, 11657, 12855, 6542, 6611, 4101, 2953, 2723, 1313, 1393; %(X0(6))
190988, 182134, 120311, 60621, 44094, 31392, 53312, 71792, 50007, 44888, 24092, 27284; %((X0(7))
72365, 46900, 21759, 11886, 15735, 8157, 18336, 11826, 6160, 5761, 3251, 3537]; %(X0(8))
[n,m]=size(x0); % Constitutes a matrix
AGO=cumsum(A);
T=1;
x1=zeros(n,m+T);
for i=1:n
for j=1:m
for k=1:j
x1(i,j)=x1(i,j)+x0(i,k);% The original data is accumulated once to get xi(1)
end
end
end
for k=1:(m-1)
Z(k)=(AGO(k)+AGO(k+1))/2; %Z(i) generates a sequence for the immediate mean of xi(1)
end
x11=x1(:,1:m);
X=x1(:,2:m)';% intercept matrix
Yn = A; %Yn is a constant term vector
Yn(1)=[];% starts from the second number, ie x(2), x(3)...
Yn=Yn';
%Yn=A(:,2:m)';
B=[-Z',X];
C=((B'*B)\(B'*Yn))';% is established by formula GM(1,n) model
a=C(1);
b=C(:,2:n+1);
F=[];
F(1)=A(1);
u=zeros(1,m);
for i=1:m
for j=1:n
u(i)=u(i)+(b(j)*x11(j,i));
end
end
for k=2:m
F(k)=(A(1)-u(k-1)/a)/exp(a*(k-1))+u(k-1)/a;
end
G=[];
G(1)=A(1);
for k=2:m
G(k)=F(k)-F(k-1);% Difference between the two to restore the original sequence, get the predicted data
end
