How to Compute Singular Value Decomposition

Singular Value Decomposition is a mathematical method for decomposing a matrix into three different, separate and informative new matrices. When you are finished with decomposing your original matrix A, you shall be in possession of a matrix with the eigenvectors of At(A) in its columns, a matrix with the eigenvectors of t(A)A in its columns and a matrix with diagonals composed of the singular values of A. By this process, you also will know the rank of A.

Things You'll Need

  • Mathematical or statistical software
Show More

Instructions

    • 1

      Compute t(A). t(A) means the transpose of matrix A; it is performed by switching the rows and columns for each index.

    • 2

      Compute t(A)A by matrix multiplication.

    • 3

      Calculate the eigenvalues of t(A)A. If you do not know how to do this, refer to the Resources section for a guide.

    • 4

      Take the square root of the eigenvalues and put them in descending order. These values are called the singular values of A.

    • 5

      Create the diagonal matrix S with the singular values in descending order being its elements.

    • 6

      Compute S-1. S-1 is the inverse of S. If you do not know how to calculate it, refer to the Resources section for a guide.

    • 7

      Calculate the eigenvectors for t(A)A.

    • 8

      Create a new matrix, V, using the eigenvectors you just computed. The eigenvectors should be the columns of V.

    • 9

      Compute t(V).

    • 10

      Calculate U = AVS-1 by matrix multiplication.

    • 11

      Get the final singular value decomposition by writing A as A = USt(V).

Learnify Hub © www.0685.com All Rights Reserved