Computing A(:, m)*B(m, :) without a for loop

Provided I have two matrices A and B of dimensions NxM and MxN I want to compute a matrix C of dimensions NxNxM such that each NxN slice is the result of the prduct A(:, m)*B(m, :). I know this can be obtained in the following way:

for m=1:M
    C(:, :, m) = A(:, m)*B(m, :);
end

Is there a way of obtaining the same result without using a for loop (with the aim of making the computation faster)?

Answer 1

[Matt J]

Ar=reshape(A,N,1,[]);
Br=reshape(B.',1,N,[]);

C=Ar.*Br;

Answer 2

[madhan ravi]

C(:, :, 1:M) = A(:, 1:M)*B(1:M, :);

[Nicolas Perez]

The output of A(:, 1:M)*B(1:M, :); is the sum of C(:, :, 1:M) along the third dimension (an NxN matrix). Thus, you get the "Assignment has fewer non-singleton rhs dimensions than non-singleton subscripts" error.