MATLAB: Matrix must be square

coefficient correlasimatrix manipulation

function Proses_CC_Callback(hObject, eventdata, handles)
% hObject    handle to Proses_CC (see GCBO)
% eventdata  reserved - to be defined in a future version of MATLAB
% handles    structure with handles and user data (see GUIDATA)
load data_x x;
load data_i i;
load data_p p;
load data_l l;
xbar=mean(x);
ibar=mean(i);
pbar=mean(p);
lbar=mean(l);
C1=sum(x-xbar)*(i-ibar)/(sum((x-xbar)^2)*sum(i-ibar)^2)^1.5;
C2=sum(x-xbar)*(p-pbar)/(sum((x-xbar)^2)*sum(p-pbar)^2)^1.5;
C3=sum(x-xbar)*(l-lbar)/(sum((x-xbar)^2)*sum(l-lbar)^2)^1.5;
%C1=(sum((x-xbar)*(i-ibar)))/((sum(x-xbar)^2)*(sum(i-ibar)^2)^1.5);
%C2=(sum((x-xbar)*(p-pbar)))/((sum(x-xbar)^2)*(sum(p-pbar)^2)^1.5);
%C3=(sum((x-xbar)*(l-lbar)))/((sum(x-xbar)^2)*(sum(l-lbar)^2)^1.5);
minimum = min([C1 C2 C3]);
if (C1 == minimum)
    class = 'temuireng'
else if (C2 == minimum)
        class = 'temuputih'
    else (C3 == minimum)
        class = 'temulawak'
    end
    set(handles.minimum, 'string',[ 'rimpang:' , num2str(class)]);
end
can help me? in the command window appears:
??? Error using ==> mpower
Matrix must be square.
Error in ==>
One_Minus_CC_new>Proses_CC_Callback at 267
C1=sum(x-xbar)*(i-ibar)/(sum((x-xbar)^2)*sum(i-ibar)^2)^1.5;

Best Answer

Did you mean to take the element-by-element power of the arrays? Look at the difference:
x = magic(2) % A 2-by-2 matrix, for example.
x^2 % No dot, equivalent to x*x, need to have square matrix.
x.^2 % Note the dot (.)  Need not be a square matrix.
So in your code you use (^2) instead of (.^2), is that what you meant to do?
Related Question