3

How to obtain sum of square of eigenvalues without finding eigenvalues of a matrix?

Rodrigo de Azevedo
  • 18,977
  • 5
  • 36
  • 95
Ka-Wa Yip
  • 830
  • 7
  • 17

3 Answers3

9

The eigenvalues of $A^2$ are the squares of the eigenvalues of $A$. The sum of the eigenvalues of any matrix (with algebraic multiplicity) is the trace. So the sum of the squares of the eigenvalues of $A$ (with algebraic multiplicity) is the trace of $A^2$.

Ian
  • 93,998
  • 3
  • 73
  • 140
  • Thanks. But why are the eigenvalues of $A^2$ are the squares of the eigenvalues of $A$? – Ka-Wa Yip Oct 04 '15 at 02:12
  • 1
    @Elessarr If $\lambda$ is an eigenvalue of $A$ with eigenvector $x$ then $A^2x=A(Ax)=A(\lambda x)=\lambda Ax=\lambda^2 x$. Ensuring the multiplicities agree is a somewhat more difficult problem. – Ian Oct 04 '15 at 02:15
1

Trace of a Matrix is the sum of its Eigenvalues

Proof that the Trace of a Matrix is the sum of its Eigenvalues

R.N
  • 4,228
  • 12
  • 23
0

For any $A\in \mathbb{R}^{n\times n}$, we have $$\sum_{i=1}^n |\lambda_i|^2 = \min_{\det(S)\neq 0} \|S^{-1}AS\|_F$$ $\lambda_i$ are the eigenvalues of $A$.

Henry Davii
  • 405
  • 6