An eigenvalue of a T:VVT:V\to V is a scalar λF\lambda\in\mathbb{F} such that there exists a nonzero vector vVv\in V with

T(v)=λv.T(v)=\lambda v.

A vector v0v\neq 0 satisfying this equation is an , and all such vectors (together with 00) form the for λ\lambda. If VV is finite-dimensional, the eigenvalues in F\mathbb F are precisely the roots in F\mathbb F of the .

Examples
  • For a diagonal matrix diag(d1,,dn)\operatorname{diag}(d_1,\dots,d_n), the eigenvalues are d1,,dnd_1,\dots,d_n.
  • For the projection P(x,y)=(x,0)P(x,y)=(x,0) on R2\mathbb{R}^2, the eigenvalues are 11 and 00.
  • On a nonzero vector space, the identity operator II has 11 as its only eigenvalue.