Nonnegative Perron vector for a symmetric nonnegative matrix #
A real symmetric entrywise-nonnegative matrix has a nonnegative unit maximizer of the
Rayleigh quotient, and that maximizer is an eigenvector for lambdaMax.
theorem
BollobasNikiforov.continuous_dotProduct_mulVec
{n : Type u_1}
[Fintype n]
{B : Matrix n n ℝ}
:
Continuous fun (x : n → ℝ) => x ⬝ᵥ B.mulVec x
theorem
BollobasNikiforov.exists_isMaxOn_dotProduct_mulVec
{n : Type u_1}
[Fintype n]
{B : Matrix n n ℝ}
(_hB : B.IsHermitian)
[Nonempty n]
:
SP02 (existence). The Rayleigh form of a symmetric matrix attains a maximum on the Euclidean unit sphere.
theorem
BollobasNikiforov.exists_nonneg_isMaxOn_dotProduct_mulVec
{n : Type u_1}
[Fintype n]
{B : Matrix n n ℝ}
(hB : B.IsHermitian)
(hnn : ∀ (i j : n), 0 ≤ B i j)
[Nonempty n]
:
SP02. If B is also entrywise nonnegative, some maximizer is nonnegative.
theorem
BollobasNikiforov.exists_nonneg_eigenvector_lambdaMax
{n : Type u_1}
[Fintype n]
[DecidableEq n]
{B : Matrix n n ℝ}
(hB : B.IsHermitian)
(hnn : ∀ (i j : n), 0 ≤ B i j)
[Nonempty n]
:
SP03. A symmetric entrywise-nonnegative matrix has a nonnegative unit eigenvector
for lambdaMax.