Error using * Matrix dimensions must agree.

when I run my code I get an error message but If i change fac with 5 for example everything runs as expected. Any idea of what can cause this?

fac is a scalar with holds the value of the element in row j column i divided by piv

Error using  * 
Incorrect dimensions for matrix
multiplication. Check that the
number of columns in the first
matrix matches the number of
rows in the second matrix. To
perform elementwise
multiplication, use '.*'.

Error in LU_CALC (line 15)
        A(j,:)=A(j,:)-fac*A(i,:); 

function [L_Matrix,U_Matrix] = LU_CALC(A)
%LU_CALC Summary of this function goes here
%   Detailed explanation goes here
m = size(A,1);
L_Matrix = eye(m);

for i=1:m
    for j=i+1:m
        piv=A(i:i);
        if piv==0
            disp("Error!");
            break;
        end
        fac=A(j:i)./piv;
        A(j,:)=A(j,:)-fac*A(i,:);
    end
end

U_Matrix = A;
end

Answer 1

[James Tursa]

Change this

piv=A(i:i);

to this

piv=A(i,i);

And change this

fac=A(j:i)./piv;

to this

fac=A(j,i)./piv;

[Ahmad Agbaria]

worked, coud you kindly explain why?

As I read matlab treats scalars as 1 by 1 matix, so what was the problem here?

[James Tursa]

A(i,j) is the (i,j)'th element of A

A(i:j) is a subset of A using linear indexing into A based on the i:j indexing. This is completely different from A(i,j)

Let's consider a 3x2 matrix just for example. So A is 3x2.

A(1,1) is the A11 element

A(1:1) is also (concidently) the A11 element because the indexing 1:1 produces the scalar index 1

A(2,1) is the A21 element

A(2:1) is an empty matrix. The indexing 2:1 means a vector starting at 2, increasing by 1, and ending at 1. Since 2 is greater than 1, the vector 2:1 is actually empty, and when using an empty vector for indexing into A you get an empty result.

E.g., A(2,3) is the A23 element, but A(2:3) is a vector containing the 2nd and 3rd elements of A which happen to be A21 and A31, not at all what you wanted.

[Ahmad Agbaria]

Thanks!