3.3 The minimum polynomial

  • Proposition 3.5. Let \(A\in M_n(\F )\). Then there is a monic polynomial \(p\in \F [x]\) such that \(p(A)=0\).

    Similarly, if \(\phi \in L(V)\) is a linear operator on a finite-dimensional vector space over \(\F \) then there is a monic polynomial \(p\in \F [x]\) with \(p(\phi )=0\).

  • Definition. A minimum polynomial for \(\phi \in L(V)\), \(V\) a vector space over \(\F \) is a monic polynomial \(p\in \F [x]\) of minimum degree with \(p(\phi )=0\): thus, if \(r\in \F [x]\) has \(r(\phi )=0\) and \(\deg r< \deg p\), then \(r=0\).

    Similarly, a minimum polynomial for \(A\in M_n(\F )\) is a monic polynomial \(p\) of least degree with \(p(A)=0\).

  • Theorem 3.6. Let \(\phi \in L(V)\) be a linear operator on a finite-dimensional vector space over a field \(\F \). Then \(\phi \) has a unique minimum polynomial.

    Similarly, any \(A\in M_n(\F )\) has a unique minimum polynomial.

    We denote these by \(m_{\phi }\) and \(m_A\) respectively.

  • Proposition 3.7. Let \(\phi \in L(V)\) be a linear operator on a finite-dimensional vector space over \(\F \) and \(p\in \F [x]\).

    Then \(p(\phi )=0\) if and only if \(m_{\phi }\) divides \(p\), that is, there is \(s\in \F [x]\) such that \(p=sm_{\phi }\).