P. 229 of Theorem List - Metamath Proof Explorer

Type

Label

Description

Statement

11.7  The characteristic polynomial

According to Wikipedia ("Characteristic polynomial", 31-Jul-2019, https://en.wikipedia.org/wiki/Characteristic_polynomial): "In linear algebra, the characteristic polynomial of a square matrix is a polynomial which is invariant under matrix similarity and has the eigenvalues as roots. It has the determinant and the trace of the matrix as coefficients.". Based on the definition of the characteristic polynomial of a square matrix (df-chpmat 22802) the eigenvalues and corresponding eigenvectors can be defined.

11.7.1  Definition and basic properties

The characteristic polynomial of a matrix 𝐴 is the determinant of the characteristic matrix of 𝐴: (𝑡𝐼 − 𝐴).

Syntax

cchpmat 22801

Extend class notation with the characteristic polynomial.

class CharPlyMat

Definition

df-chpmat 22802*

Define the characteristic polynomial of a square matrix. According to Wikipedia ("Characteristic polynomial", 31-Jul-2019, https://en.wikipedia.org/wiki/Characteristic_polynomial): "The characteristic polynomial of [an n x n matrix] A, denoted by p_A(t), is the polynomial defined by p_A ( t ) = det ( t I - A ) where I denotes the n-by-n identity matrix.". In addition, however, the underlying ring must be commutative, see definition in [Lang], p. 561: " Let k be a commutative ring ... Let M be any n x n matrix in k ... We define the characteristic polynomial P_M(t) to be the determinant det ( t I_n - M ) where I_n is the unit n x n matrix." To be more precise, the matrices A and I on the right hand side are matrices with coefficients of a polynomial ring. Therefore, the original matrix A over a given commutative ring must be transformed into corresponding matrices over the polynomial ring over the given ring. (Contributed by AV, 2-Aug-2019.)

⊢ CharPlyMat = (𝑛 ∈ Fin, 𝑟 ∈ V ↦ (𝑚 ∈ (Base‘(𝑛 Mat 𝑟)) ↦ ((𝑛 maDet (Poly₁‘𝑟))‘(((var₁‘𝑟)( ·_𝑠 ‘(𝑛 Mat (Poly₁‘𝑟)))(1_r‘(𝑛 Mat (Poly₁‘𝑟))))(-_g‘(𝑛 Mat (Poly₁‘𝑟)))((𝑛 matToPolyMat 𝑟)‘𝑚)))))

Theorem

chmatcl 22803

Closure of the characteristic matrix of a matrix. (Contributed by AV, 25-Oct-2019.) (Proof shortened by AV, 29-Nov-2019.)

⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝐵 = (Base‘𝐴)    &   ⊢ 𝑃 = (Poly₁‘𝑅)    &   ⊢ 𝑌 = (𝑁 Mat 𝑃)    &   ⊢ 𝑋 = (var₁‘𝑅)    &   ⊢ 𝑇 = (𝑁 matToPolyMat 𝑅)    &   ⊢ − = (-_g‘𝑌)    &   ⊢ · = ( ·_𝑠 ‘𝑌)    &   ⊢ 1 = (1_r‘𝑌)    &   ⊢ 𝐻 = ((𝑋 · 1 ) − (𝑇‘𝑀))    ⇒   ⊢ ((𝑁 ∈ Fin ∧ 𝑅 ∈ Ring ∧ 𝑀 ∈ 𝐵) → 𝐻 ∈ (Base‘𝑌))

Theorem

chmatval 22804

The entries of the characteristic matrix of a matrix. (Contributed by AV, 2-Aug-2019.) (Proof shortened by AV, 10-Dec-2019.)

⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝐵 = (Base‘𝐴)    &   ⊢ 𝑃 = (Poly₁‘𝑅)    &   ⊢ 𝑌 = (𝑁 Mat 𝑃)    &   ⊢ 𝑋 = (var₁‘𝑅)    &   ⊢ 𝑇 = (𝑁 matToPolyMat 𝑅)    &   ⊢ − = (-_g‘𝑌)    &   ⊢ · = ( ·_𝑠 ‘𝑌)    &   ⊢ 1 = (1_r‘𝑌)    &   ⊢ 𝐻 = ((𝑋 · 1 ) − (𝑇‘𝑀))    &   ⊢ ∼ = (-_g‘𝑃)    &   ⊢ 0 = (0_g‘𝑃)    ⇒   ⊢ (((𝑁 ∈ Fin ∧ 𝑅 ∈ Ring ∧ 𝑀 ∈ 𝐵) ∧ (𝐼 ∈ 𝑁 ∧ 𝐽 ∈ 𝑁)) → (𝐼𝐻𝐽) = if(𝐼 = 𝐽, (𝑋 ∼ (𝐼(𝑇‘𝑀)𝐽)), ( 0 ∼ (𝐼(𝑇‘𝑀)𝐽))))

Theorem

chpmatfval 22805*

Value of the characteristic polynomial function. (Contributed by AV, 2-Aug-2019.)

⊢ 𝐶 = (𝑁 CharPlyMat 𝑅)    &   ⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝐵 = (Base‘𝐴)    &   ⊢ 𝑃 = (Poly₁‘𝑅)    &   ⊢ 𝑌 = (𝑁 Mat 𝑃)    &   ⊢ 𝐷 = (𝑁 maDet 𝑃)    &   ⊢ − = (-_g‘𝑌)    &   ⊢ 𝑋 = (var₁‘𝑅)    &   ⊢ · = ( ·_𝑠 ‘𝑌)    &   ⊢ 𝑇 = (𝑁 matToPolyMat 𝑅)    &   ⊢ 1 = (1_r‘𝑌)    ⇒   ⊢ ((𝑁 ∈ Fin ∧ 𝑅 ∈ 𝑉) → 𝐶 = (𝑚 ∈ 𝐵 ↦ (𝐷‘((𝑋 · 1 ) − (𝑇‘𝑚)))))

Theorem

chpmatval 22806

The characteristic polynomial of a (square) matrix (expressed with a determinant). (Contributed by AV, 2-Aug-2019.)

⊢ 𝐶 = (𝑁 CharPlyMat 𝑅)    &   ⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝐵 = (Base‘𝐴)    &   ⊢ 𝑃 = (Poly₁‘𝑅)    &   ⊢ 𝑌 = (𝑁 Mat 𝑃)    &   ⊢ 𝐷 = (𝑁 maDet 𝑃)    &   ⊢ − = (-_g‘𝑌)    &   ⊢ 𝑋 = (var₁‘𝑅)    &   ⊢ · = ( ·_𝑠 ‘𝑌)    &   ⊢ 𝑇 = (𝑁 matToPolyMat 𝑅)    &   ⊢ 1 = (1_r‘𝑌)    ⇒   ⊢ ((𝑁 ∈ Fin ∧ 𝑅 ∈ 𝑉 ∧ 𝑀 ∈ 𝐵) → (𝐶‘𝑀) = (𝐷‘((𝑋 · 1 ) − (𝑇‘𝑀))))

Theorem

chpmatply1 22807

The characteristic polynomial of a (square) matrix over a commutative ring is a polynomial, see also the following remark in [Lang], p. 561: "[the characteristic polynomial] is an element of k[t]". (Contributed by AV, 2-Aug-2019.) (Proof shortened by AV, 29-Nov-2019.)

⊢ 𝐶 = (𝑁 CharPlyMat 𝑅)    &   ⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝐵 = (Base‘𝐴)    &   ⊢ 𝑃 = (Poly₁‘𝑅)    &   ⊢ 𝐸 = (Base‘𝑃)    ⇒   ⊢ ((𝑁 ∈ Fin ∧ 𝑅 ∈ CRing ∧ 𝑀 ∈ 𝐵) → (𝐶‘𝑀) ∈ 𝐸)

Theorem

chpmatval2 22808*

The characteristic polynomial of a (square) matrix (expressed with the Leibnitz formula for the determinant). (Contributed by AV, 2-Aug-2019.)

⊢ 𝐶 = (𝑁 CharPlyMat 𝑅)    &   ⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝐵 = (Base‘𝐴)    &   ⊢ 𝑃 = (Poly₁‘𝑅)    &   ⊢ 𝑌 = (𝑁 Mat 𝑃)    &   ⊢ − = (-_g‘𝑌)    &   ⊢ 𝑋 = (var₁‘𝑅)    &   ⊢ · = ( ·_𝑠 ‘𝑌)    &   ⊢ 𝑇 = (𝑁 matToPolyMat 𝑅)    &   ⊢ 1 = (1_r‘𝑌)    &   ⊢ 𝐺 = (SymGrp‘𝑁)    &   ⊢ 𝐻 = (Base‘𝐺)    &   ⊢ 𝑍 = (ℤRHom‘𝑃)    &   ⊢ 𝑆 = (pmSgn‘𝑁)    &   ⊢ 𝑈 = (mulGrp‘𝑃)    &   ⊢ × = (._r‘𝑃)    ⇒   ⊢ ((𝑁 ∈ Fin ∧ 𝑅 ∈ Ring ∧ 𝑀 ∈ 𝐵) → (𝐶‘𝑀) = (𝑃 Σ_g (𝑝 ∈ 𝐻 ↦ (((𝑍 ∘ 𝑆)‘𝑝) × (𝑈 Σ_g (𝑥 ∈ 𝑁 ↦ ((𝑝‘𝑥)((𝑋 · 1 ) − (𝑇‘𝑀))𝑥)))))))

Theorem

chpmat0d 22809

The characteristic polynomial of the empty matrix. (Contributed by AV, 6-Aug-2019.)

⊢ 𝐶 = (∅ CharPlyMat 𝑅)    ⇒   ⊢ (𝑅 ∈ Ring → (𝐶‘∅) = (1_r‘(Poly₁‘𝑅)))

Theorem

chpmat1dlem 22810

Lemma for chpmat1d 22811. (Contributed by AV, 7-Aug-2019.)

⊢ 𝐶 = (𝑁 CharPlyMat 𝑅)    &   ⊢ 𝑃 = (Poly₁‘𝑅)    &   ⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝐵 = (Base‘𝐴)    &   ⊢ 𝑋 = (var₁‘𝑅)    &   ⊢ − = (-_g‘𝑃)    &   ⊢ 𝑆 = (algSc‘𝑃)    &   ⊢ 𝐺 = (𝑁 Mat 𝑃)    &   ⊢ 𝑇 = (𝑁 matToPolyMat 𝑅)    ⇒   ⊢ ((𝑅 ∈ Ring ∧ (𝑁 = {𝐼} ∧ 𝐼 ∈ 𝑉) ∧ 𝑀 ∈ 𝐵) → (𝐼((𝑋( ·_𝑠 ‘𝐺)(1_r‘𝐺))(-_g‘𝐺)(𝑇‘𝑀))𝐼) = (𝑋 − (𝑆‘(𝐼𝑀𝐼))))

Theorem

chpmat1d 22811

The characteristic polynomial of a matrix with dimension 1. (Contributed by AV, 7-Aug-2019.)

⊢ 𝐶 = (𝑁 CharPlyMat 𝑅)    &   ⊢ 𝑃 = (Poly₁‘𝑅)    &   ⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝐵 = (Base‘𝐴)    &   ⊢ 𝑋 = (var₁‘𝑅)    &   ⊢ − = (-_g‘𝑃)    &   ⊢ 𝑆 = (algSc‘𝑃)    ⇒   ⊢ ((𝑅 ∈ CRing ∧ (𝑁 = {𝐼} ∧ 𝐼 ∈ 𝑉) ∧ 𝑀 ∈ 𝐵) → (𝐶‘𝑀) = (𝑋 − (𝑆‘(𝐼𝑀𝐼))))

Theorem

chpdmatlem0 22812

Lemma 0 for chpdmat 22816. (Contributed by AV, 18-Aug-2019.)

⊢ 𝐶 = (𝑁 CharPlyMat 𝑅)    &   ⊢ 𝑃 = (Poly₁‘𝑅)    &   ⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝑆 = (algSc‘𝑃)    &   ⊢ 𝐵 = (Base‘𝐴)    &   ⊢ 𝑋 = (var₁‘𝑅)    &   ⊢ 0 = (0_g‘𝑅)    &   ⊢ 𝐺 = (mulGrp‘𝑃)    &   ⊢ − = (-_g‘𝑃)    &   ⊢ 𝑄 = (𝑁 Mat 𝑃)    &   ⊢ 1 = (1_r‘𝑄)    &   ⊢ · = ( ·_𝑠 ‘𝑄)    ⇒   ⊢ ((𝑁 ∈ Fin ∧ 𝑅 ∈ Ring) → (𝑋 · 1 ) ∈ (Base‘𝑄))

Theorem

chpdmatlem1 22813

Lemma 1 for chpdmat 22816. (Contributed by AV, 18-Aug-2019.)

⊢ 𝐶 = (𝑁 CharPlyMat 𝑅)    &   ⊢ 𝑃 = (Poly₁‘𝑅)    &   ⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝑆 = (algSc‘𝑃)    &   ⊢ 𝐵 = (Base‘𝐴)    &   ⊢ 𝑋 = (var₁‘𝑅)    &   ⊢ 0 = (0_g‘𝑅)    &   ⊢ 𝐺 = (mulGrp‘𝑃)    &   ⊢ − = (-_g‘𝑃)    &   ⊢ 𝑄 = (𝑁 Mat 𝑃)    &   ⊢ 1 = (1_r‘𝑄)    &   ⊢ · = ( ·_𝑠 ‘𝑄)    &   ⊢ 𝑍 = (-_g‘𝑄)    &   ⊢ 𝑇 = (𝑁 matToPolyMat 𝑅)    ⇒   ⊢ ((𝑁 ∈ Fin ∧ 𝑅 ∈ Ring ∧ 𝑀 ∈ 𝐵) → ((𝑋 · 1 )𝑍(𝑇‘𝑀)) ∈ (Base‘𝑄))

Theorem

chpdmatlem2 22814

Lemma 2 for chpdmat 22816. (Contributed by AV, 18-Aug-2019.)

⊢ 𝐶 = (𝑁 CharPlyMat 𝑅)    &   ⊢ 𝑃 = (Poly₁‘𝑅)    &   ⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝑆 = (algSc‘𝑃)    &   ⊢ 𝐵 = (Base‘𝐴)    &   ⊢ 𝑋 = (var₁‘𝑅)    &   ⊢ 0 = (0_g‘𝑅)    &   ⊢ 𝐺 = (mulGrp‘𝑃)    &   ⊢ − = (-_g‘𝑃)    &   ⊢ 𝑄 = (𝑁 Mat 𝑃)    &   ⊢ 1 = (1_r‘𝑄)    &   ⊢ · = ( ·_𝑠 ‘𝑄)    &   ⊢ 𝑍 = (-_g‘𝑄)    &   ⊢ 𝑇 = (𝑁 matToPolyMat 𝑅)    ⇒   ⊢ ((((((𝑁 ∈ Fin ∧ 𝑅 ∈ Ring ∧ 𝑀 ∈ 𝐵) ∧ 𝑖 ∈ 𝑁) ∧ 𝑗 ∈ 𝑁) ∧ 𝑖 ≠ 𝑗) ∧ (𝑖𝑀𝑗) = 0 ) → (𝑖((𝑋 · 1 )𝑍(𝑇‘𝑀))𝑗) = (0_g‘𝑃))

Theorem

chpdmatlem3 22815

Lemma 3 for chpdmat 22816. (Contributed by AV, 18-Aug-2019.)

⊢ 𝐶 = (𝑁 CharPlyMat 𝑅)    &   ⊢ 𝑃 = (Poly₁‘𝑅)    &   ⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝑆 = (algSc‘𝑃)    &   ⊢ 𝐵 = (Base‘𝐴)    &   ⊢ 𝑋 = (var₁‘𝑅)    &   ⊢ 0 = (0_g‘𝑅)    &   ⊢ 𝐺 = (mulGrp‘𝑃)    &   ⊢ − = (-_g‘𝑃)    &   ⊢ 𝑄 = (𝑁 Mat 𝑃)    &   ⊢ 1 = (1_r‘𝑄)    &   ⊢ · = ( ·_𝑠 ‘𝑄)    &   ⊢ 𝑍 = (-_g‘𝑄)    &   ⊢ 𝑇 = (𝑁 matToPolyMat 𝑅)    ⇒   ⊢ (((𝑁 ∈ Fin ∧ 𝑅 ∈ Ring ∧ 𝑀 ∈ 𝐵) ∧ 𝐾 ∈ 𝑁) → (𝐾((𝑋 · 1 )𝑍(𝑇‘𝑀))𝐾) = (𝑋 − (𝑆‘(𝐾𝑀𝐾))))

Theorem

chpdmat 22816*

The characteristic polynomial of a diagonal matrix. (Contributed by AV, 18-Aug-2019.) (Proof shortened by AV, 21-Nov-2019.)

⊢ 𝐶 = (𝑁 CharPlyMat 𝑅)    &   ⊢ 𝑃 = (Poly₁‘𝑅)    &   ⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝑆 = (algSc‘𝑃)    &   ⊢ 𝐵 = (Base‘𝐴)    &   ⊢ 𝑋 = (var₁‘𝑅)    &   ⊢ 0 = (0_g‘𝑅)    &   ⊢ 𝐺 = (mulGrp‘𝑃)    &   ⊢ − = (-_g‘𝑃)    ⇒   ⊢ (((𝑁 ∈ Fin ∧ 𝑅 ∈ CRing ∧ 𝑀 ∈ 𝐵) ∧ ∀𝑖 ∈ 𝑁 ∀𝑗 ∈ 𝑁 (𝑖 ≠ 𝑗 → (𝑖𝑀𝑗) = 0 )) → (𝐶‘𝑀) = (𝐺 Σ_g (𝑘 ∈ 𝑁 ↦ (𝑋 − (𝑆‘(𝑘𝑀𝑘))))))

Theorem

chpscmat 22817*

The characteristic polynomial of a (nonempty!) scalar matrix. (Contributed by AV, 21-Aug-2019.)

⊢ 𝐶 = (𝑁 CharPlyMat 𝑅)    &   ⊢ 𝑃 = (Poly₁‘𝑅)    &   ⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝑋 = (var₁‘𝑅)    &   ⊢ 𝐺 = (mulGrp‘𝑃)    &   ⊢ ↑ = (._g‘𝐺)    &   ⊢ 𝐷 = {𝑚 ∈ (Base‘𝐴) ∣ ∃𝑐 ∈ (Base‘𝑅)∀𝑖 ∈ 𝑁 ∀𝑗 ∈ 𝑁 (𝑖𝑚𝑗) = if(𝑖 = 𝑗, 𝑐, (0_g‘𝑅))}    &   ⊢ 𝑆 = (algSc‘𝑃)    &   ⊢ − = (-_g‘𝑃)    ⇒   ⊢ (((𝑁 ∈ Fin ∧ 𝑅 ∈ CRing) ∧ (𝑀 ∈ 𝐷 ∧ 𝐼 ∈ 𝑁 ∧ ∀𝑛 ∈ 𝑁 (𝑛𝑀𝑛) = 𝐸)) → (𝐶‘𝑀) = ((♯‘𝑁) ↑ (𝑋 − (𝑆‘𝐸))))

Theorem

chpscmat0 22818*

The characteristic polynomial of a (nonempty!) scalar matrix, expressed with its diagonal element. (Contributed by AV, 21-Aug-2019.)

⊢ 𝐶 = (𝑁 CharPlyMat 𝑅)    &   ⊢ 𝑃 = (Poly₁‘𝑅)    &   ⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝑋 = (var₁‘𝑅)    &   ⊢ 𝐺 = (mulGrp‘𝑃)    &   ⊢ ↑ = (._g‘𝐺)    &   ⊢ 𝐷 = {𝑚 ∈ (Base‘𝐴) ∣ ∃𝑐 ∈ (Base‘𝑅)∀𝑖 ∈ 𝑁 ∀𝑗 ∈ 𝑁 (𝑖𝑚𝑗) = if(𝑖 = 𝑗, 𝑐, (0_g‘𝑅))}    &   ⊢ 𝑆 = (algSc‘𝑃)    &   ⊢ − = (-_g‘𝑃)    ⇒   ⊢ (((𝑁 ∈ Fin ∧ 𝑅 ∈ CRing) ∧ (𝑀 ∈ 𝐷 ∧ 𝐼 ∈ 𝑁 ∧ ∀𝑛 ∈ 𝑁 (𝑛𝑀𝑛) = (𝐼𝑀𝐼))) → (𝐶‘𝑀) = ((♯‘𝑁) ↑ (𝑋 − (𝑆‘(𝐼𝑀𝐼)))))

Theorem

chpscmatgsumbin 22819*

The characteristic polynomial of a (nonempty!) scalar matrix, expressed as finite group sum of binomials. (Contributed by AV, 2-Sep-2019.)

⊢ 𝐶 = (𝑁 CharPlyMat 𝑅)    &   ⊢ 𝑃 = (Poly₁‘𝑅)    &   ⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝑋 = (var₁‘𝑅)    &   ⊢ 𝐺 = (mulGrp‘𝑃)    &   ⊢ ↑ = (._g‘𝐺)    &   ⊢ 𝐷 = {𝑚 ∈ (Base‘𝐴) ∣ ∃𝑐 ∈ (Base‘𝑅)∀𝑖 ∈ 𝑁 ∀𝑗 ∈ 𝑁 (𝑖𝑚𝑗) = if(𝑖 = 𝑗, 𝑐, (0_g‘𝑅))}    &   ⊢ 𝑆 = (algSc‘𝑃)    &   ⊢ − = (-_g‘𝑃)    &   ⊢ 𝐹 = (._g‘𝑃)    &   ⊢ 𝐻 = (mulGrp‘𝑅)    &   ⊢ 𝐸 = (._g‘𝐻)    &   ⊢ 𝐼 = (inv_g‘𝑅)    &   ⊢ · = ( ·_𝑠 ‘𝑃)    ⇒   ⊢ (((𝑁 ∈ Fin ∧ 𝑅 ∈ CRing) ∧ (𝑀 ∈ 𝐷 ∧ 𝐽 ∈ 𝑁 ∧ ∀𝑛 ∈ 𝑁 (𝑛𝑀𝑛) = (𝐽𝑀𝐽))) → (𝐶‘𝑀) = (𝑃 Σ_g (𝑙 ∈ (0...(♯‘𝑁)) ↦ (((♯‘𝑁)C𝑙)𝐹((((♯‘𝑁) − 𝑙)𝐸(𝐼‘(𝐽𝑀𝐽))) · (𝑙 ↑ 𝑋))))))

Theorem

chpscmatgsummon 22820*

The characteristic polynomial of a (nonempty!) scalar matrix, expressed as finite group sum of scaled monomials. (Contributed by AV, 2-Sep-2019.)

⊢ 𝐶 = (𝑁 CharPlyMat 𝑅)    &   ⊢ 𝑃 = (Poly₁‘𝑅)    &   ⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝑋 = (var₁‘𝑅)    &   ⊢ 𝐺 = (mulGrp‘𝑃)    &   ⊢ ↑ = (._g‘𝐺)    &   ⊢ 𝐷 = {𝑚 ∈ (Base‘𝐴) ∣ ∃𝑐 ∈ (Base‘𝑅)∀𝑖 ∈ 𝑁 ∀𝑗 ∈ 𝑁 (𝑖𝑚𝑗) = if(𝑖 = 𝑗, 𝑐, (0_g‘𝑅))}    &   ⊢ 𝑆 = (algSc‘𝑃)    &   ⊢ − = (-_g‘𝑃)    &   ⊢ 𝐹 = (._g‘𝑃)    &   ⊢ 𝐻 = (mulGrp‘𝑅)    &   ⊢ 𝐸 = (._g‘𝐻)    &   ⊢ 𝐼 = (inv_g‘𝑅)    &   ⊢ · = ( ·_𝑠 ‘𝑃)    &   ⊢ 𝑍 = (._g‘𝑅)    ⇒   ⊢ (((𝑁 ∈ Fin ∧ 𝑅 ∈ CRing) ∧ (𝑀 ∈ 𝐷 ∧ 𝐽 ∈ 𝑁 ∧ ∀𝑛 ∈ 𝑁 (𝑛𝑀𝑛) = (𝐽𝑀𝐽))) → (𝐶‘𝑀) = (𝑃 Σ_g (𝑙 ∈ (0...(♯‘𝑁)) ↦ ((((♯‘𝑁)C𝑙)𝑍(((♯‘𝑁) − 𝑙)𝐸(𝐼‘(𝐽𝑀𝐽)))) · (𝑙 ↑ 𝑋)))))

Theorem

chp0mat 22821

The characteristic polynomial of the zero matrix. (Contributed by AV, 18-Aug-2019.)

⊢ 𝐶 = (𝑁 CharPlyMat 𝑅)    &   ⊢ 𝑃 = (Poly₁‘𝑅)    &   ⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝑋 = (var₁‘𝑅)    &   ⊢ 𝐺 = (mulGrp‘𝑃)    &   ⊢ ↑ = (._g‘𝐺)    &   ⊢ 0 = (0_g‘𝐴)    ⇒   ⊢ ((𝑁 ∈ Fin ∧ 𝑅 ∈ CRing) → (𝐶‘ 0 ) = ((♯‘𝑁) ↑ 𝑋))

Theorem

chpidmat 22822

The characteristic polynomial of the identity matrix. (Contributed by AV, 19-Aug-2019.)

⊢ 𝐶 = (𝑁 CharPlyMat 𝑅)    &   ⊢ 𝑃 = (Poly₁‘𝑅)    &   ⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝑋 = (var₁‘𝑅)    &   ⊢ 𝐺 = (mulGrp‘𝑃)    &   ⊢ ↑ = (._g‘𝐺)    &   ⊢ 𝐼 = (1_r‘𝐴)    &   ⊢ 𝑆 = (algSc‘𝑃)    &   ⊢ 1 = (1_r‘𝑅)    &   ⊢ − = (-_g‘𝑃)    ⇒   ⊢ ((𝑁 ∈ Fin ∧ 𝑅 ∈ CRing) → (𝐶‘𝐼) = ((♯‘𝑁) ↑ (𝑋 − (𝑆‘ 1 ))))

Theorem

chmaidscmat 22823

The characteristic polynomial of a matrix multiplied with the identity matrix is a scalar matrix. (Contributed by AV, 30-Oct-2019.) (Revised by AV, 5-Jul-2022.)

⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝐵 = (Base‘𝐴)    &   ⊢ 𝐶 = (𝑁 CharPlyMat 𝑅)    &   ⊢ 𝑃 = (Poly₁‘𝑅)    &   ⊢ 𝐸 = (Base‘𝑃)    &   ⊢ 𝑌 = (𝑁 Mat 𝑃)    &   ⊢ 𝐾 = (Base‘𝑌)    &   ⊢ · = ( ·_𝑠 ‘𝑌)    &   ⊢ 1 = (1_r‘𝑌)    &   ⊢ 𝑆 = (𝑁 ScMat 𝑃)    ⇒   ⊢ ((𝑁 ∈ Fin ∧ 𝑅 ∈ CRing ∧ 𝑀 ∈ 𝐵) → ((𝐶‘𝑀) · 1 ) ∈ 𝑆)

11.7.2  The characteristic factor function G

In this subsection the function 𝐺 = (𝑛 ∈ ℕ₀ ↦ if(𝑛 = 0, ( 0 − ((𝑇‘𝑀) × (𝑇‘(𝑏‘0)))), if(𝑛 = (𝑠 + 1), (𝑇‘(𝑏‘𝑠)), if((𝑠 + 1) < 𝑛, 0 , ((𝑇‘(𝑏‘(𝑛 − 1))) − ((𝑇‘𝑀) × (𝑇‘(𝑏‘𝑛)))))))) is discussed. This function is involved in the representation of the product of the characteristic matrix of a given matrix and its adjunct as an infinite sum, see cpmadugsum 22853. Therefore, this function is called "characteristic factor function" (in short "chfacf") in the following. It plays an important role in the proof of the Cayley-Hamilton theorem, see cayhamlem1 22841, cayhamlem3 22862 and cayhamlem4 22863.

Theorem

fvmptnn04if 22824*

The function values of a mapping from the nonnegative integers with four distinct cases. (Contributed by AV, 10-Nov-2019.)

⊢ 𝐺 = (𝑛 ∈ ℕ₀ ↦ if(𝑛 = 0, 𝐴, if(𝑛 = 𝑆, 𝐶, if(𝑆 < 𝑛, 𝐷, 𝐵))))    &   ⊢ (𝜑 → 𝑆 ∈ ℕ)    &   ⊢ (𝜑 → 𝑁 ∈ ℕ₀)    &   ⊢ (𝜑 → 𝑌 ∈ 𝑉)    &   ⊢ ((𝜑 ∧ 𝑁 = 0) → 𝑌 = ⦋𝑁 / 𝑛⦌𝐴)    &   ⊢ ((𝜑 ∧ 0 < 𝑁 ∧ 𝑁 < 𝑆) → 𝑌 = ⦋𝑁 / 𝑛⦌𝐵)    &   ⊢ ((𝜑 ∧ 𝑁 = 𝑆) → 𝑌 = ⦋𝑁 / 𝑛⦌𝐶)    &   ⊢ ((𝜑 ∧ 𝑆 < 𝑁) → 𝑌 = ⦋𝑁 / 𝑛⦌𝐷)    ⇒   ⊢ (𝜑 → (𝐺‘𝑁) = 𝑌)

Theorem

fvmptnn04ifa 22825*

The function value of a mapping from the nonnegative integers with four distinct cases for the first case. (Contributed by AV, 10-Nov-2019.)

⊢ 𝐺 = (𝑛 ∈ ℕ₀ ↦ if(𝑛 = 0, 𝐴, if(𝑛 = 𝑆, 𝐶, if(𝑆 < 𝑛, 𝐷, 𝐵))))    &   ⊢ (𝜑 → 𝑆 ∈ ℕ)    &   ⊢ (𝜑 → 𝑁 ∈ ℕ₀)    ⇒   ⊢ ((𝜑 ∧ 𝑁 = 0 ∧ ⦋𝑁 / 𝑛⦌𝐴 ∈ 𝑉) → (𝐺‘𝑁) = ⦋𝑁 / 𝑛⦌𝐴)

Theorem

fvmptnn04ifb 22826*

The function value of a mapping from the nonnegative integers with four distinct cases for the second case. (Contributed by AV, 10-Nov-2019.)

⊢ 𝐺 = (𝑛 ∈ ℕ₀ ↦ if(𝑛 = 0, 𝐴, if(𝑛 = 𝑆, 𝐶, if(𝑆 < 𝑛, 𝐷, 𝐵))))    &   ⊢ (𝜑 → 𝑆 ∈ ℕ)    &   ⊢ (𝜑 → 𝑁 ∈ ℕ₀)    ⇒   ⊢ ((𝜑 ∧ (0 < 𝑁 ∧ 𝑁 < 𝑆) ∧ ⦋𝑁 / 𝑛⦌𝐵 ∈ 𝑉) → (𝐺‘𝑁) = ⦋𝑁 / 𝑛⦌𝐵)

Theorem

fvmptnn04ifc 22827*

The function value of a mapping from the nonnegative integers with four distinct cases for the third case. (Contributed by AV, 10-Nov-2019.)

⊢ 𝐺 = (𝑛 ∈ ℕ₀ ↦ if(𝑛 = 0, 𝐴, if(𝑛 = 𝑆, 𝐶, if(𝑆 < 𝑛, 𝐷, 𝐵))))    &   ⊢ (𝜑 → 𝑆 ∈ ℕ)    &   ⊢ (𝜑 → 𝑁 ∈ ℕ₀)    ⇒   ⊢ ((𝜑 ∧ 𝑁 = 𝑆 ∧ ⦋𝑁 / 𝑛⦌𝐶 ∈ 𝑉) → (𝐺‘𝑁) = ⦋𝑁 / 𝑛⦌𝐶)

Theorem

fvmptnn04ifd 22828*

The function value of a mapping from the nonnegative integers with four distinct cases for the forth case. (Contributed by AV, 10-Nov-2019.)

⊢ 𝐺 = (𝑛 ∈ ℕ₀ ↦ if(𝑛 = 0, 𝐴, if(𝑛 = 𝑆, 𝐶, if(𝑆 < 𝑛, 𝐷, 𝐵))))    &   ⊢ (𝜑 → 𝑆 ∈ ℕ)    &   ⊢ (𝜑 → 𝑁 ∈ ℕ₀)    ⇒   ⊢ ((𝜑 ∧ 𝑆 < 𝑁 ∧ ⦋𝑁 / 𝑛⦌𝐷 ∈ 𝑉) → (𝐺‘𝑁) = ⦋𝑁 / 𝑛⦌𝐷)

Theorem

chfacfisf 22829*

The "characteristic factor function" is a function from the nonnegative integers to polynomial matrices. (Contributed by AV, 8-Nov-2019.)

⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝐵 = (Base‘𝐴)    &   ⊢ 𝑃 = (Poly₁‘𝑅)    &   ⊢ 𝑌 = (𝑁 Mat 𝑃)    &   ⊢ × = (._r‘𝑌)    &   ⊢ − = (-_g‘𝑌)    &   ⊢ 0 = (0_g‘𝑌)    &   ⊢ 𝑇 = (𝑁 matToPolyMat 𝑅)    &   ⊢ 𝐺 = (𝑛 ∈ ℕ₀ ↦ if(𝑛 = 0, ( 0 − ((𝑇‘𝑀) × (𝑇‘(𝑏‘0)))), if(𝑛 = (𝑠 + 1), (𝑇‘(𝑏‘𝑠)), if((𝑠 + 1) < 𝑛, 0 , ((𝑇‘(𝑏‘(𝑛 − 1))) − ((𝑇‘𝑀) × (𝑇‘(𝑏‘𝑛))))))))    ⇒   ⊢ (((𝑁 ∈ Fin ∧ 𝑅 ∈ Ring ∧ 𝑀 ∈ 𝐵) ∧ (𝑠 ∈ ℕ ∧ 𝑏 ∈ (𝐵 ↑_m (0...𝑠)))) → 𝐺:ℕ₀⟶(Base‘𝑌))

Theorem

chfacfisfcpmat 22830*

The "characteristic factor function" is a function from the nonnegative integers to constant polynomial matrices. (Contributed by AV, 19-Nov-2019.)

⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝐵 = (Base‘𝐴)    &   ⊢ 𝑃 = (Poly₁‘𝑅)    &   ⊢ 𝑌 = (𝑁 Mat 𝑃)    &   ⊢ × = (._r‘𝑌)    &   ⊢ − = (-_g‘𝑌)    &   ⊢ 0 = (0_g‘𝑌)    &   ⊢ 𝑇 = (𝑁 matToPolyMat 𝑅)    &   ⊢ 𝐺 = (𝑛 ∈ ℕ₀ ↦ if(𝑛 = 0, ( 0 − ((𝑇‘𝑀) × (𝑇‘(𝑏‘0)))), if(𝑛 = (𝑠 + 1), (𝑇‘(𝑏‘𝑠)), if((𝑠 + 1) < 𝑛, 0 , ((𝑇‘(𝑏‘(𝑛 − 1))) − ((𝑇‘𝑀) × (𝑇‘(𝑏‘𝑛))))))))    &   ⊢ 𝑆 = (𝑁 ConstPolyMat 𝑅)    ⇒   ⊢ (((𝑁 ∈ Fin ∧ 𝑅 ∈ Ring ∧ 𝑀 ∈ 𝐵) ∧ (𝑠 ∈ ℕ ∧ 𝑏 ∈ (𝐵 ↑_m (0...𝑠)))) → 𝐺:ℕ₀⟶𝑆)

Theorem

chfacffsupp 22831*

The "characteristic factor function" is finitely supported. (Contributed by AV, 20-Nov-2019.) (Proof shortened by AV, 23-Dec-2019.)

⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝐵 = (Base‘𝐴)    &   ⊢ 𝑃 = (Poly₁‘𝑅)    &   ⊢ 𝑌 = (𝑁 Mat 𝑃)    &   ⊢ × = (._r‘𝑌)    &   ⊢ − = (-_g‘𝑌)    &   ⊢ 0 = (0_g‘𝑌)    &   ⊢ 𝑇 = (𝑁 matToPolyMat 𝑅)    &   ⊢ 𝐺 = (𝑛 ∈ ℕ₀ ↦ if(𝑛 = 0, ( 0 − ((𝑇‘𝑀) × (𝑇‘(𝑏‘0)))), if(𝑛 = (𝑠 + 1), (𝑇‘(𝑏‘𝑠)), if((𝑠 + 1) < 𝑛, 0 , ((𝑇‘(𝑏‘(𝑛 − 1))) − ((𝑇‘𝑀) × (𝑇‘(𝑏‘𝑛))))))))    ⇒   ⊢ (((𝑁 ∈ Fin ∧ 𝑅 ∈ CRing ∧ 𝑀 ∈ 𝐵) ∧ (𝑠 ∈ ℕ ∧ 𝑏 ∈ (𝐵 ↑_m (0...𝑠)))) → 𝐺 finSupp (0_g‘𝑌))

Theorem

chfacfscmulcl 22832*

Closure of a scaled value of the "characteristic factor function". (Contributed by AV, 9-Nov-2019.)

⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝐵 = (Base‘𝐴)    &   ⊢ 𝑃 = (Poly₁‘𝑅)    &   ⊢ 𝑌 = (𝑁 Mat 𝑃)    &   ⊢ × = (._r‘𝑌)    &   ⊢ − = (-_g‘𝑌)    &   ⊢ 0 = (0_g‘𝑌)    &   ⊢ 𝑇 = (𝑁 matToPolyMat 𝑅)    &   ⊢ 𝐺 = (𝑛 ∈ ℕ₀ ↦ if(𝑛 = 0, ( 0 − ((𝑇‘𝑀) × (𝑇‘(𝑏‘0)))), if(𝑛 = (𝑠 + 1), (𝑇‘(𝑏‘𝑠)), if((𝑠 + 1) < 𝑛, 0 , ((𝑇‘(𝑏‘(𝑛 − 1))) − ((𝑇‘𝑀) × (𝑇‘(𝑏‘𝑛))))))))    &   ⊢ 𝑋 = (var₁‘𝑅)    &   ⊢ · = ( ·_𝑠 ‘𝑌)    &   ⊢ ↑ = (._g‘(mulGrp‘𝑃))    ⇒   ⊢ (((𝑁 ∈ Fin ∧ 𝑅 ∈ CRing ∧ 𝑀 ∈ 𝐵) ∧ (𝑠 ∈ ℕ ∧ 𝑏 ∈ (𝐵 ↑_m (0...𝑠))) ∧ 𝐾 ∈ ℕ₀) → ((𝐾 ↑ 𝑋) · (𝐺‘𝐾)) ∈ (Base‘𝑌))

Theorem

chfacfscmul0 22833*

A scaled value of the "characteristic factor function" is zero almost always. (Contributed by AV, 9-Nov-2019.)

⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝐵 = (Base‘𝐴)    &   ⊢ 𝑃 = (Poly₁‘𝑅)    &   ⊢ 𝑌 = (𝑁 Mat 𝑃)    &   ⊢ × = (._r‘𝑌)    &   ⊢ − = (-_g‘𝑌)    &   ⊢ 0 = (0_g‘𝑌)    &   ⊢ 𝑇 = (𝑁 matToPolyMat 𝑅)    &   ⊢ 𝐺 = (𝑛 ∈ ℕ₀ ↦ if(𝑛 = 0, ( 0 − ((𝑇‘𝑀) × (𝑇‘(𝑏‘0)))), if(𝑛 = (𝑠 + 1), (𝑇‘(𝑏‘𝑠)), if((𝑠 + 1) < 𝑛, 0 , ((𝑇‘(𝑏‘(𝑛 − 1))) − ((𝑇‘𝑀) × (𝑇‘(𝑏‘𝑛))))))))    &   ⊢ 𝑋 = (var₁‘𝑅)    &   ⊢ · = ( ·_𝑠 ‘𝑌)    &   ⊢ ↑ = (._g‘(mulGrp‘𝑃))    ⇒   ⊢ (((𝑁 ∈ Fin ∧ 𝑅 ∈ CRing ∧ 𝑀 ∈ 𝐵) ∧ (𝑠 ∈ ℕ ∧ 𝑏 ∈ (𝐵 ↑_m (0...𝑠))) ∧ 𝐾 ∈ (ℤ_≥‘(𝑠 + 2))) → ((𝐾 ↑ 𝑋) · (𝐺‘𝐾)) = 0 )

Theorem

chfacfscmulfsupp 22834*

A mapping of scaled values of the "characteristic factor function" is finitely supported. (Contributed by AV, 8-Nov-2019.)

⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝐵 = (Base‘𝐴)    &   ⊢ 𝑃 = (Poly₁‘𝑅)    &   ⊢ 𝑌 = (𝑁 Mat 𝑃)    &   ⊢ × = (._r‘𝑌)    &   ⊢ − = (-_g‘𝑌)    &   ⊢ 0 = (0_g‘𝑌)    &   ⊢ 𝑇 = (𝑁 matToPolyMat 𝑅)    &   ⊢ 𝐺 = (𝑛 ∈ ℕ₀ ↦ if(𝑛 = 0, ( 0 − ((𝑇‘𝑀) × (𝑇‘(𝑏‘0)))), if(𝑛 = (𝑠 + 1), (𝑇‘(𝑏‘𝑠)), if((𝑠 + 1) < 𝑛, 0 , ((𝑇‘(𝑏‘(𝑛 − 1))) − ((𝑇‘𝑀) × (𝑇‘(𝑏‘𝑛))))))))    &   ⊢ 𝑋 = (var₁‘𝑅)    &   ⊢ · = ( ·_𝑠 ‘𝑌)    &   ⊢ ↑ = (._g‘(mulGrp‘𝑃))    ⇒   ⊢ (((𝑁 ∈ Fin ∧ 𝑅 ∈ CRing ∧ 𝑀 ∈ 𝐵) ∧ (𝑠 ∈ ℕ ∧ 𝑏 ∈ (𝐵 ↑_m (0...𝑠)))) → (𝑖 ∈ ℕ₀ ↦ ((𝑖 ↑ 𝑋) · (𝐺‘𝑖))) finSupp 0 )

Theorem

chfacfscmulgsum 22835*

Breaking up a sum of values of the "characteristic factor function" scaled by a polynomial. (Contributed by AV, 9-Nov-2019.)

⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝐵 = (Base‘𝐴)    &   ⊢ 𝑃 = (Poly₁‘𝑅)    &   ⊢ 𝑌 = (𝑁 Mat 𝑃)    &   ⊢ × = (._r‘𝑌)    &   ⊢ − = (-_g‘𝑌)    &   ⊢ 0 = (0_g‘𝑌)    &   ⊢ 𝑇 = (𝑁 matToPolyMat 𝑅)    &   ⊢ 𝐺 = (𝑛 ∈ ℕ₀ ↦ if(𝑛 = 0, ( 0 − ((𝑇‘𝑀) × (𝑇‘(𝑏‘0)))), if(𝑛 = (𝑠 + 1), (𝑇‘(𝑏‘𝑠)), if((𝑠 + 1) < 𝑛, 0 , ((𝑇‘(𝑏‘(𝑛 − 1))) − ((𝑇‘𝑀) × (𝑇‘(𝑏‘𝑛))))))))    &   ⊢ 𝑋 = (var₁‘𝑅)    &   ⊢ · = ( ·_𝑠 ‘𝑌)    &   ⊢ ↑ = (._g‘(mulGrp‘𝑃))    &   ⊢ + = (+_g‘𝑌)    ⇒   ⊢ (((𝑁 ∈ Fin ∧ 𝑅 ∈ CRing ∧ 𝑀 ∈ 𝐵) ∧ (𝑠 ∈ ℕ ∧ 𝑏 ∈ (𝐵 ↑_m (0...𝑠)))) → (𝑌 Σ_g (𝑖 ∈ ℕ₀ ↦ ((𝑖 ↑ 𝑋) · (𝐺‘𝑖)))) = ((𝑌 Σ_g (𝑖 ∈ (1...𝑠) ↦ ((𝑖 ↑ 𝑋) · ((𝑇‘(𝑏‘(𝑖 − 1))) − ((𝑇‘𝑀) × (𝑇‘(𝑏‘𝑖))))))) + ((((𝑠 + 1) ↑ 𝑋) · (𝑇‘(𝑏‘𝑠))) − ((𝑇‘𝑀) × (𝑇‘(𝑏‘0))))))

Theorem

chfacfpmmulcl 22836*

Closure of the value of the "characteristic factor function" multiplied with a constant polynomial matrix. (Contributed by AV, 23-Nov-2019.)

⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝐵 = (Base‘𝐴)    &   ⊢ 𝑃 = (Poly₁‘𝑅)    &   ⊢ 𝑌 = (𝑁 Mat 𝑃)    &   ⊢ × = (._r‘𝑌)    &   ⊢ − = (-_g‘𝑌)    &   ⊢ 0 = (0_g‘𝑌)    &   ⊢ 𝑇 = (𝑁 matToPolyMat 𝑅)    &   ⊢ 𝐺 = (𝑛 ∈ ℕ₀ ↦ if(𝑛 = 0, ( 0 − ((𝑇‘𝑀) × (𝑇‘(𝑏‘0)))), if(𝑛 = (𝑠 + 1), (𝑇‘(𝑏‘𝑠)), if((𝑠 + 1) < 𝑛, 0 , ((𝑇‘(𝑏‘(𝑛 − 1))) − ((𝑇‘𝑀) × (𝑇‘(𝑏‘𝑛))))))))    &   ⊢ ↑ = (._g‘(mulGrp‘𝑌))    ⇒   ⊢ (((𝑁 ∈ Fin ∧ 𝑅 ∈ CRing ∧ 𝑀 ∈ 𝐵) ∧ (𝑠 ∈ ℕ ∧ 𝑏 ∈ (𝐵 ↑_m (0...𝑠))) ∧ 𝐾 ∈ ℕ₀) → ((𝐾 ↑ (𝑇‘𝑀)) × (𝐺‘𝐾)) ∈ (Base‘𝑌))

Theorem

chfacfpmmul0 22837*

The value of the "characteristic factor function" multiplied with a constant polynomial matrix is zero almost always. (Contributed by AV, 23-Nov-2019.)

⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝐵 = (Base‘𝐴)    &   ⊢ 𝑃 = (Poly₁‘𝑅)    &   ⊢ 𝑌 = (𝑁 Mat 𝑃)    &   ⊢ × = (._r‘𝑌)    &   ⊢ − = (-_g‘𝑌)    &   ⊢ 0 = (0_g‘𝑌)    &   ⊢ 𝑇 = (𝑁 matToPolyMat 𝑅)    &   ⊢ 𝐺 = (𝑛 ∈ ℕ₀ ↦ if(𝑛 = 0, ( 0 − ((𝑇‘𝑀) × (𝑇‘(𝑏‘0)))), if(𝑛 = (𝑠 + 1), (𝑇‘(𝑏‘𝑠)), if((𝑠 + 1) < 𝑛, 0 , ((𝑇‘(𝑏‘(𝑛 − 1))) − ((𝑇‘𝑀) × (𝑇‘(𝑏‘𝑛))))))))    &   ⊢ ↑ = (._g‘(mulGrp‘𝑌))    ⇒   ⊢ (((𝑁 ∈ Fin ∧ 𝑅 ∈ CRing ∧ 𝑀 ∈ 𝐵) ∧ (𝑠 ∈ ℕ ∧ 𝑏 ∈ (𝐵 ↑_m (0...𝑠))) ∧ 𝐾 ∈ (ℤ_≥‘(𝑠 + 2))) → ((𝐾 ↑ (𝑇‘𝑀)) × (𝐺‘𝐾)) = 0 )

Theorem

chfacfpmmulfsupp 22838*

A mapping of values of the "characteristic factor function" multiplied with a constant polynomial matrix is finitely supported. (Contributed by AV, 23-Nov-2019.)

⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝐵 = (Base‘𝐴)    &   ⊢ 𝑃 = (Poly₁‘𝑅)    &   ⊢ 𝑌 = (𝑁 Mat 𝑃)    &   ⊢ × = (._r‘𝑌)    &   ⊢ − = (-_g‘𝑌)    &   ⊢ 0 = (0_g‘𝑌)    &   ⊢ 𝑇 = (𝑁 matToPolyMat 𝑅)    &   ⊢ 𝐺 = (𝑛 ∈ ℕ₀ ↦ if(𝑛 = 0, ( 0 − ((𝑇‘𝑀) × (𝑇‘(𝑏‘0)))), if(𝑛 = (𝑠 + 1), (𝑇‘(𝑏‘𝑠)), if((𝑠 + 1) < 𝑛, 0 , ((𝑇‘(𝑏‘(𝑛 − 1))) − ((𝑇‘𝑀) × (𝑇‘(𝑏‘𝑛))))))))    &   ⊢ ↑ = (._g‘(mulGrp‘𝑌))    ⇒   ⊢ (((𝑁 ∈ Fin ∧ 𝑅 ∈ CRing ∧ 𝑀 ∈ 𝐵) ∧ (𝑠 ∈ ℕ ∧ 𝑏 ∈ (𝐵 ↑_m (0...𝑠)))) → (𝑖 ∈ ℕ₀ ↦ ((𝑖 ↑ (𝑇‘𝑀)) × (𝐺‘𝑖))) finSupp 0 )

Theorem

chfacfpmmulgsum 22839*

Breaking up a sum of values of the "characteristic factor function" multiplied with a constant polynomial matrix. (Contributed by AV, 23-Nov-2019.)

⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝐵 = (Base‘𝐴)    &   ⊢ 𝑃 = (Poly₁‘𝑅)    &   ⊢ 𝑌 = (𝑁 Mat 𝑃)    &   ⊢ × = (._r‘𝑌)    &   ⊢ − = (-_g‘𝑌)    &   ⊢ 0 = (0_g‘𝑌)    &   ⊢ 𝑇 = (𝑁 matToPolyMat 𝑅)    &   ⊢ 𝐺 = (𝑛 ∈ ℕ₀ ↦ if(𝑛 = 0, ( 0 − ((𝑇‘𝑀) × (𝑇‘(𝑏‘0)))), if(𝑛 = (𝑠 + 1), (𝑇‘(𝑏‘𝑠)), if((𝑠 + 1) < 𝑛, 0 , ((𝑇‘(𝑏‘(𝑛 − 1))) − ((𝑇‘𝑀) × (𝑇‘(𝑏‘𝑛))))))))    &   ⊢ ↑ = (._g‘(mulGrp‘𝑌))    &   ⊢ + = (+_g‘𝑌)    ⇒   ⊢ (((𝑁 ∈ Fin ∧ 𝑅 ∈ CRing ∧ 𝑀 ∈ 𝐵) ∧ (𝑠 ∈ ℕ ∧ 𝑏 ∈ (𝐵 ↑_m (0...𝑠)))) → (𝑌 Σ_g (𝑖 ∈ ℕ₀ ↦ ((𝑖 ↑ (𝑇‘𝑀)) × (𝐺‘𝑖)))) = ((𝑌 Σ_g (𝑖 ∈ (1...𝑠) ↦ ((𝑖 ↑ (𝑇‘𝑀)) × ((𝑇‘(𝑏‘(𝑖 − 1))) − ((𝑇‘𝑀) × (𝑇‘(𝑏‘𝑖))))))) + ((((𝑠 + 1) ↑ (𝑇‘𝑀)) × (𝑇‘(𝑏‘𝑠))) − ((𝑇‘𝑀) × (𝑇‘(𝑏‘0))))))

Theorem

chfacfpmmulgsum2 22840*

Breaking up a sum of values of the "characteristic factor function" multiplied with a constant polynomial matrix. (Contributed by AV, 23-Nov-2019.)

⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝐵 = (Base‘𝐴)    &   ⊢ 𝑃 = (Poly₁‘𝑅)    &   ⊢ 𝑌 = (𝑁 Mat 𝑃)    &   ⊢ × = (._r‘𝑌)    &   ⊢ − = (-_g‘𝑌)    &   ⊢ 0 = (0_g‘𝑌)    &   ⊢ 𝑇 = (𝑁 matToPolyMat 𝑅)    &   ⊢ 𝐺 = (𝑛 ∈ ℕ₀ ↦ if(𝑛 = 0, ( 0 − ((𝑇‘𝑀) × (𝑇‘(𝑏‘0)))), if(𝑛 = (𝑠 + 1), (𝑇‘(𝑏‘𝑠)), if((𝑠 + 1) < 𝑛, 0 , ((𝑇‘(𝑏‘(𝑛 − 1))) − ((𝑇‘𝑀) × (𝑇‘(𝑏‘𝑛))))))))    &   ⊢ ↑ = (._g‘(mulGrp‘𝑌))    &   ⊢ + = (+_g‘𝑌)    ⇒   ⊢ (((𝑁 ∈ Fin ∧ 𝑅 ∈ CRing ∧ 𝑀 ∈ 𝐵) ∧ (𝑠 ∈ ℕ ∧ 𝑏 ∈ (𝐵 ↑_m (0...𝑠)))) → (𝑌 Σ_g (𝑖 ∈ ℕ₀ ↦ ((𝑖 ↑ (𝑇‘𝑀)) × (𝐺‘𝑖)))) = ((𝑌 Σ_g (𝑖 ∈ (1...𝑠) ↦ (((𝑖 ↑ (𝑇‘𝑀)) × (𝑇‘(𝑏‘(𝑖 − 1)))) − (((𝑖 + 1) ↑ (𝑇‘𝑀)) × (𝑇‘(𝑏‘𝑖)))))) + ((((𝑠 + 1) ↑ (𝑇‘𝑀)) × (𝑇‘(𝑏‘𝑠))) − ((𝑇‘𝑀) × (𝑇‘(𝑏‘0))))))

Theorem

cayhamlem1 22841*

Lemma 1 for cayleyhamilton 22865. (Contributed by AV, 11-Nov-2019.)

⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝐵 = (Base‘𝐴)    &   ⊢ 𝑃 = (Poly₁‘𝑅)    &   ⊢ 𝑌 = (𝑁 Mat 𝑃)    &   ⊢ × = (._r‘𝑌)    &   ⊢ − = (-_g‘𝑌)    &   ⊢ 0 = (0_g‘𝑌)    &   ⊢ 𝑇 = (𝑁 matToPolyMat 𝑅)    &   ⊢ 𝐺 = (𝑛 ∈ ℕ₀ ↦ if(𝑛 = 0, ( 0 − ((𝑇‘𝑀) × (𝑇‘(𝑏‘0)))), if(𝑛 = (𝑠 + 1), (𝑇‘(𝑏‘𝑠)), if((𝑠 + 1) < 𝑛, 0 , ((𝑇‘(𝑏‘(𝑛 − 1))) − ((𝑇‘𝑀) × (𝑇‘(𝑏‘𝑛))))))))    &   ⊢ ↑ = (._g‘(mulGrp‘𝑌))    ⇒   ⊢ (((𝑁 ∈ Fin ∧ 𝑅 ∈ CRing ∧ 𝑀 ∈ 𝐵) ∧ (𝑠 ∈ ℕ ∧ 𝑏 ∈ (𝐵 ↑_m (0...𝑠)))) → (𝑌 Σ_g (𝑖 ∈ ℕ₀ ↦ ((𝑖 ↑ (𝑇‘𝑀)) × (𝐺‘𝑖)))) = 0 )

11.7.3  The Cayley-Hamilton theorem

In this section, a direct algebraic proof for the Cayley-Hamilton theorem is provided, according to Wikipedia ("Cayley-Hamilton theorem", 09-Nov-2019, https://en.wikipedia.org/wiki/Cayley%E2%80%93Hamilton_theorem, section "A direct algebraic proof" (this approach is also used for proving Lemma 1.9 in [Hefferon] p. 427):

"This proof uses just the kind of objects needed to formulate the Cayley-Hamilton theorem: matrices with polynomials as entries. The matrix (t * I_n - A) whose determinant is the characteristic polynomial of A is such a matrix, and since polynomials [over a commutative ring] form a commutative ring, it has an adjugate

(1) B = adj(t * I_n - A) .

Then, according to the right-hand fundamental relation of the adjugate, one has

(2) ( t * I_n - A ) x B = det(t * I_n - A) x I_n = p(t) * I_n .

Since B is also a matrix with polynomials in t as entries, one can, for each i, collect the coefficients of t^i in each entry to form a matrix B_i of numbers, such that one has

(3) B = sum_{i = 0 to (n-1)} t^i B_i .

(The way the entries of B are defined makes clear that no powers higher than t^(n-1) occur). While this looks like a polynomial with matrices as coefficients, we shall not consider such a notion; it is just a way to write a matrix with polynomial entries as a linear combination of n constant matrices, and the coefficient t^i has been written to the left of the matrix to stress this point of view.

Now, one can expand the matrix product in our equation by bilinearity

(4) p(t) * I_n = ( t * I_n - A ) x B
= ( t * I_n - A ) x sum_{i = 0 to (n-1)} t^i * B_i
= sum_{i = 0 to (n-1)} t * I_n x t^i B_i - sum_{i = 0 to (n-1)} A * t^i B_i
= sum_{i = 0 to (n-1)} t^(i+1) * B_i - sum_{i = 0 to (n-1)} t^i * A x B_i
= t^n B_n-1 + sum_{i = 1 to (n-1)} t^i * ( B_i-1 - A x B_i ) - A x B₀ .

Writing

(5) p(t) I_n = t^n * I_n + t^(n-1) * c_(n-1) x I_n + ... + t * c₁ I_n + c₀ I_n ,

one obtains an equality of two matrices with polynomial entries, written as linear combinations of constant matrices with powers of t as coefficients. Such an equality can hold only if in any matrix position the entry that is multiplied by a given power t^i is the same on both sides; it follows that the constant matrices with coefficient t^i in both expressions must be equal. Writing these equations then for i from n down to 0, one finds

(6) B_n-1 = I_n , B_i-1 - A x B_i = c_i * I_n for 1 <= i <= n-1 , - A x B₀ = c₀ * I_n .

Finally, multiply the equation of the coefficients of t^i from the left by A^i, and sum up:

(7) A^n B_n-1 + sum_{i = 1 to (n-1)} ( A^i x B_i-1 - A^(i+1) x B_i ) - A x B₀ = A^n + c_n-1 * A^(n-1) + ... + c₁ * A + c₀ * I_n .

The left-hand sides form a telescoping sum and cancel completely; the right-hand sides add up to p(A):

(8) 0 = p(A) .

This completes the proof."

To formalize this approach, the steps mentioned in Wikipedia must be elaborated in more detail.

The first step is to formalize the preliminaries and the objective of the theorem. In Wikipedia, the Cayley-Hamilton theorem is stated as follows: "... the Cayley-Hamilton theorem ... states that every square matrix over a commutative ring ... satisfies its own characteristic equation." Or in more detail: "If A is a given n x n matrix and I_n is the n x n identity matrix, then the characteristic polynomial of A is defined as p(t) = det(t * I_n - A), where det is the determinant operation and t is a variable for a scalar element of the base ring. Since the entries of the matrix (t * I_n - A) are (linear or constant) polynomials in t, the determinant is also an n-th order monic polynomial in t. The Cayley-Hamilton theorem states that if one defines an analogous matrix equation, p(A), consisting of the replacement of the scalar eigenvalues t with the matrix A, then this polynomial in the matrix A results in the zero matrix,

p(A) = 0.

The powers of A, obtained by substitution from powers of t, are defined by repeated matrix multiplication; the constant term of p(t) gives a multiple of the power A^0, which is defined as the identity matrix. The theorem allows A^n to be expressed as a linear combination of the lower matrix powers of A. When the ring is a field, the Cayley-Hamilton theorem is equivalent to the statement that the minimal polynomial of a square matrix divides its characteristic polynomial."

Actually, the definition of the characteristic polynomial of a square matrix requires some attention. According to df-chpmat 22802, the characteristic polynomial of an 𝑁 x 𝑁 matrix 𝑀 over a ring 𝑅 is defined as

((𝑁 CharPlyMat 𝑅)‘𝑀) = (𝐷‘((𝑋 · 1 ) − (𝑇‘𝑀))))

where 𝐷 = (𝑁 maDet 𝑃) is the function mapping an 𝑁 x 𝑁 matrix over the polynomial ring over the ring 𝑅 to its determinant, 𝑋 = (var₁‘𝑅) is the variable of the polynomials over 𝑅, 1 is the 𝑁 x 𝑁 identity matrix as matrix over the polynomial ring over the ring 𝑅 (not the 𝑁 x 𝑁 identity matrix of the matrices over the ring 𝑅!) and (𝑇‘𝑀) = ((𝑁 matToPolyMat 𝑅)‘𝑀) is the matrix 𝑀 over a ring 𝑅 transformed into a constant matrix over the polynomial ring over the ring 𝑅. Thus · is the multiplication of a polynomial matrix with a "scalar", i.e. a polynomial (see chpmatval 22806).

By this definition, it is ensured that ((𝑋 · 1 ) − (𝑇‘𝑀)), the matrix whose determinant is the characteristic polynomial of the matrix 𝑀, is actually a matrix over the polynomial ring over the ring 𝑅, as stated in Wikipedia ("matrix with polynomials as entries"). This matrix is called the characteristic matrix of matrix 𝑀 (see Wikipedia "Polynomial matrix", 16-Nov-2019, https://en.wikipedia.org/wiki/Polynomial_matrix 22806). Following the notation in Wikipedia, we denote the characteristic polynomial of the matrix 𝑀, formally defined by ((𝑁 CharPlyMat 𝑅)‘𝑀) as "p(M)" in the comments. The characteristic matrix ((𝑋 · 1 ) − (𝑇‘𝑀)) will be denoted by "c(M)", so that "p(M) = det( c(M) )".

After the preliminaries are clarified, the objective of the Cayley-Hamilton theorem must be considered. As described in Wikipedia, the matrix 𝑀 must be "inserted" into its characteristic polynomial in an appropriate way. Since every polynomial can be represented as infinite, but finitely supported sum of monomials scaled by the corresponding coefficients (see ply1coe 22273), also the characteristic polynomial can be written in this way:

p(M) = sum_i ( p_i * M^i )

Here, * is the scalar multiplication in the algebra of the polynomials over the ring 𝑅, and the coefficients are elements of the ring 𝑅.

By this, we can "define" the insertion of the matrix M into its characteristic polynomial by "p(M) = sum( p_i * M^i)", see also cayleyhamilton1 22867. Here, * is the scalar multiplication in the algebra of the matrices over the ring 𝑅.

To prove the Cayley-Hamilton theorem, we have to show that "p(M) = 0", where 0 is the zero matrix.

In this section, the following class variables and informal identifiers (acronyms in the form "A(B)" or "A_B") are used:

class variable/ informal identifier	definiens	explanation
𝑁		An arbitrary finite set, used as dimension for matrices
𝑅		An arbitrary (commutative) ring: 𝑅 ∈ CRing
B(R)	(Base‘𝑅)	Base set of (the ring) 𝑅
𝐴	(𝑁 Mat 𝑅)	Algebra of 𝑁 x 𝑁 matrices over (the ring) 𝑅
𝐵	(Base‘𝐴)	Base set of the algebra of 𝑁 x 𝑁 matrices, i .e. the set of all 𝑁 x 𝑁 matrices
𝑀		An arbitrary 𝑁 x 𝑁 matrix
𝑃	(Poly1‘𝑅)	The algebra of polynomials over (the ring) 𝑅
B(P)	(Base‘𝑃)	Base set of the algebra of polynomials, i .e. the set of all polynomials
𝑋, XR	(var1‘𝑅)	The variable of polynomials over (the ring) 𝑅
𝑌, XA	(var1‘𝐴)	The variable of polynomials over matrices of the algebra 𝐴
↑	(.g‘(mulGrp‘𝑃))	The group exponentiation for polynomials over (the ring) 𝑅
^		Arbitrary group exponentiation
𝑈	(algSc‘𝑃)	The injection of scalars, i.e. elements of (the ring) 𝑅 into the base set of the algebra of polynomials over 𝑅
(𝑈‘𝑝), S(p)	((algSc‘𝑃)‘𝑝)	The element 𝑝 of (the ring) 𝑅 represented as polynomial over 𝑅
𝑌	(𝑁 Mat 𝑃)	Algebra of 𝑁 x 𝑁 matrices over (the polynomial ring) 𝑃 over the ring 𝑅
B(Y)	(Base‘𝑌)	Base set of the algebra of polynomial 𝑁 x 𝑁 matrices, i .e. the set of all polynomial 𝑁 x 𝑁 matrices
𝑄	(Poly1‘𝐴)	Algebra of polynomials over the ring of 𝑁 x 𝑁 matrices over the ring 𝑅
B(Q)	(Base‘𝑄)	Base set of the algebra of polynomials over the ring of 𝑁 x 𝑁 matrices over the ring 𝑅, i .e. the set of all polynomials having 𝑁 x 𝑁 matrices as coefficients
+, +	(+g‘𝑌)	The addition of polynomial matrices
−, -	(-g‘𝑌)	The subtraction of polynomial matrices
·, *Y	( ·𝑠 ‘𝑌)	The multiplication of a polynomial matrix with a scalar ( i. e. a polynomial)
*A	( ·𝑠 ‘𝐴)	The multiplication of a matrix with a scalar ( i. e. an element of the underlying ring)
*Q	( ·𝑠 ‘𝑄)	The multiplication of a polynomial over matrices with a scalar ( i. e. a matrix)
×, xY	(.r‘𝑌)	The multiplication of polynomial matrices
xA	(.r‘𝐴)	The multiplication of matrices
xQ	(.r‘𝑄)	The multiplication of polynomials over matrices
1, 1Y	(1r‘𝑌)	The identity matrix in the algebra of polynomial matrices over 𝑅
1A	(1r‘𝐴)	The identity matrix in the algebra of matrices over 𝑅
0, 0Y	(0g‘𝑌)	The zero matrix in the algebra of matrices consisting of polynomials
𝑇	(𝑁 matToPolyMat 𝑅)	The transformation of an 𝑁 x 𝑁 matrix over 𝑅 into a polynomial 𝑁 x 𝑁 matrix over 𝑅
T1(M)	(𝑇‘𝑀)	The matrix M transformed into a polynomial 𝑁 x 𝑁 matrix over 𝑅
U(M)	(𝑈‘𝑀)	The (constant) polynomial 𝑁 x 𝑁 matrix M transformed into a matrix over the ring 𝑅. Inverse function of 𝑇: (𝑇‘(𝑈‘𝑀)) = 𝑀 (see m2cpminvid2 22730 )
T2(M)	((𝑁 pMatToMatPoly 𝑅)‘𝑀)	The polynomial 𝑁 x 𝑁 matrix M transformed into a polynomial over the 𝑁 x 𝑁 matrices over 𝑅
𝐼, c(M)	((𝑋 · 1 ) − (𝑇‘𝑀))	The characteristic matrix of a matrix 𝑀, i.e. the matrix whose determinant is the characteristic polynomial of the matrix 𝑀
𝐶	(𝑁 CharPlyMat 𝑅)	The function mapping a matrix over (a ring) 𝑅 to its characteristic polynomial
𝐾, p(M)	(𝐶‘𝑀)	The characteristic polynomial of a matrix over (a ring) 𝑅
𝐻	(𝐾 · 1 )	The scalar matrix (diagonal matrix) with the characteristic polynomial of a matrix as diagonal elements
𝐽	(𝑁 maAdju 𝑃)	The function mapping a matrix consisting of polynomials to its adjugate ("matrix of cofactors")
𝑊, adj(cm(M))	(𝐽‘𝐼)	The adjugate of the characteristic matrix of the matrix 𝑀

Using this notation, we have:

1. "c(M) e. B(Y)", or 𝐼 ∈ (Base‘𝑌), see chmatcl 22803
2. "p(M) e. B(P)", or 𝐾 ∈ (Base‘𝑃), see chpmatply1 22807
3. "T(M) e. B(Y)", or (𝑇‘𝑀) ∈ (Base‘𝑌), see mat2pmatbas 22701
4. 𝐽:(Base‘𝑌)⟶(Base‘𝑌), see maduf 22616
5. "adj(cm(M)) e. B(Y)", or 𝑊 ∈ (Base‘𝑌)

Following the proof shown in Wikipedia, the following steps are performed:

1. Write 𝑊, the adjugate of the characteristic matrix, as a finite sum of scaled monomials, see pmatcollpw3fi1 22763:
   adj(cm(M)) = sum_{i=0 to s} ( X_R ^i *_Y T₁(b(i)) )
   where b(i) are matrices over the ring 𝑅, so T₁(b(i)) are constant polynomial matrices.
   This step corresponds to (3) in Wikipedia. In contrast to Wikipedia, we write 𝑊 as finite sum of not exactly determined number of summands, which may be greater than needed (including summands of value 0). This will be sufficient to provide a representation of (𝐼 × 𝑊) as infinite, but finitely supported sum, see step 3.
2. Write (𝐼 × 𝑊), the product of the characteristic matrix and its adjugate as finite sum of scaled monomials, see cpmadugsumfi 22852. This representation is obtained by replacing 𝑊 by the representation resulting from step 1. and performing calculation rules available for the associative algebra of matrices over polynomials over a commutative ring:
   cm(M) *_Y adj(cm(M)) = sum_{i=0 to s} ( X_R ^i *_Y ( T₁(b(i-1)) - T₁(M) x_Y T₁(b(i)) ) ) + X_R ^(s+1) *_Y ( T₁(b(s)) - T₁(M) x_Y T₁(b(0))
   where b(i) are matrices over 𝑅, so T₁(b(i)) are constant polynomial matrices:
   cm(M) *_Y adj(cm(M))
   = cm(M) *_Y sum_{i=0 to s} ( X_R ^i *_Y T₁(b(i)) ) [see pmatcollpw3fi1 22763 (step 1.)]
   = ( ( X_A *_Y 1_Y ) - T₁(M) ) *_Y sum_{i=0 to s} ( X_R ^i *_Y T₁(b(i)) ) [def. of cm(M)]
   = ( X_A *_Y 1_Y ) *_Y sum_{i=0 to s} ( X_R ^i *_Y T₁(b(i)) ) - T₁(M) *_Y sum_{i=0 to s} ( X_R ^i *_Y T₁(b(i)) ) [see ringsubdir 20280]
   = sum_{i=0 to s} ( X_R ^i *_Y ( T₁(b(i-1)) - T₁(M) x_Y T₁(b(i)) ) ) + X_R ^(s+1) *_Y ( T₁(b(s)) - T₁(M) x_Y T₁(b(0)) [see cpmadugsumlemF 22851]
   This step corresponds partially to (4) in Wikipedia.
3. Write (𝐼 × 𝑊) as infinite, but finitely supported sum of scaled monomials, see cpmadugsum 22853:
   cm(M) * adj(cm(M)) = sum_i ( X_R ^i *_Y G(i) )
   This representation is obtained by defining a function G for the coefficients, which we call "characteristic factor function", see chfacfisf 22829, which covers the special terms and the padding with 0. G(i) is a constant polynomial matrix (see chfacfisfcpmat 22830). This step corresponds partially to (4) in Wikipedia, with summands of value 0 added.
4. Write 𝐻 = (𝐾 · 1 ), the scalar matrix (diagonal matrix) with the characteristic polynomial of a matrix as diagonal elements, as infinite, but finitely supported sum of scaled monomials. See cpmidgsum 22843:
   p(m) *_Y I_Y = sum_i ( X_R ^i *_Y ( S(p_i) *_Y I_Y ) )
   The proof of cpmidgsum 22843 is making use of pmatcollpwscmat 22766, because 𝐻 = (𝐾 · 1 ) is a scalar/diagonal polynomial matrix with the characteristic polynomial "p(M)" as diagonal entries (since p_i is an element of the ring 𝑅, S(p_i) is a (constant) polynomial). This corresponds to (5) in Wikipedia, with summands of value 0 added.
5. Transform the sum representation of (𝐼 × 𝑊) from step 3. into polynomials over matrices:
   T₂(cm(M) * adj(cm(M))) = sum_i ( U(G(i)) *_Q X_A ^i ) [see cpmadumatpoly 22858]
   where U(G(i)) is a matrix over the ring 𝑅.
6. Transform the sum representation of 𝐻 from step 4. into polynomials over matrices:
   T₂(p(m) *_Y I_Y) = sum_i ( p_i *_A I_A ) *_Q X_A ^i ) [see cpmidpmat 22848]
7. Equate the sum representations resulting from steps 5. and 6. by using cpmadurid 22842 to obtain the equation
   sum_i ( U(G(i)) *_Q X_A ^i ) = sum_i ( p_i *_A I_A ) *_Q X_A ^i ):
   sum_i ( U(G(i)) *_Q X_A ^i )
   = T₂(cm(M) * adj(cm(M))) [see step 5.]
   = T₂(p(m) *_Y I_Y) [see cpmadurid 22842]
   = sum_i ( p_i *_A I_A ) *_Q X_A ^i ) [see step 6.]
   Note that this step is contained in the proof of chcoeffeq 22861, see step 9. This step corresponds to the conclusion from (4) and (5) in Wikipedia, with summands of value 0 added.
8. Compare the sum representations of step 7. to obtain the equations U(G(i)) = p_i *_A I_A , see chcoeffeqlem 22860. This corresponds to (6) in Wikipedia. Since the coefficients of the transformed representations and the original representations are identical, the equations of the coefficients are also valid for the original representations of steps 3. and 4.
9. Multiply the equations of the coefficients from step 8. from the left by M^i, and sum up, see chcoeffeq 22861:
   sum_i ( M^i x_A U(G(i)) ) = sum_i ( M^i x_A ( p_i *_A I_A) )
   This corresponds to (7) in Wikipedia.
10. Transform the right hand side of the equation in step 9. into an appropriate form, see cayhamlem3 22862:
    sum_i ( p_i *_A M^i )
    = sum_i ( M^i x_A ( p_i *_A I_A) ) [see cayhamlem2 22859]
    = sum_i ( M^i x_A U(G(i)) ) [see chcoeffeq 22861]
11. Apply the theorem for telescoping sums, see telgsumfz 19956, to the sum sum_i ( T₁(M)^i x_Y G(i) ), which results in an equation to 0:
    sum_i ( T₁(M)^i x_Y G(i) ) = 0_Y, see cayhamlem1 22841:
    sum_i ( T₁(M)^i x_Y G(i) )
    = sum_{i=1 to s} ( T₁(M)^i x_Y T₁(b(i-1)) - T₁(M)^(i+1) x_Y T₁(b(i)) )
    + ( T₁(M)^(s+1) x_Y T₁(b(s)) - T₁(M) x_Y T₁(b(0)) ) [see chfacfpmmulgsum2 22840]
    = ( T₁(M) x_Y T₁(b(0)) - T₁(M)^(s+1) x_Y T₁(b(s)) ) + ( T₁ M)^(s+1) x_Y T₁(b(s)) - T₁(M) x_Y T₁(b(0)) ) [see telgsumfz 19956]
    = 0_Y [see grpnpncan0 19003] This step corresponds partially to (8) in Wikipedia.
12. Since 𝑇 is a ring homomorphism (see mat2pmatrhm 22709), the left hand side of the equation in step 10. can be transformed into a representation appropriate to apply the result of step 11., see cayhamlem4 22863:
    sum_i ( p_i *_A M^i )
    = sum_i ( M^i x_A U(G(i)) ) [see cayhamlem3 22862 (step 10.)]
    = U(T₁(sum_i ( M^i x_A U(G(i)) ))) [see m2cpminvid 22728]
    = U(sum_i T₁( M^i x_A U(G(i)) )) [see gsummptmhm 19906]
    = U(sum_i ( T₁(M^i) x_Y T₁(U(G(i))) )) [see rhmmul 20456]
    = U(sum_i ( T₁(M)^i x_Y T₁(U(G(i))) )) [see mhmmulg 19082]
    = U(sum_i ( T₁(M)^i x_Y G(i) )) [see m2cpminvid2 22730 ]
    
13. Finally, combine the results of steps 11. and 12., and use the fact that 𝑇 (and therefore also its inverse 𝑈) is an injective ring homomorphism (see mat2pmatf1 22704 and mat2pmatrhm 22709) to transform the equality resulting from steps 11. and 12. into the desired equation sum_i ( p_i *_A M^i ) = 0_A , see cayleyhamilton 22865 resp. cayleyhamilton0 22864:
    sum_i ( p_i *_A M^i )
    = U(sum_i ( T₁(M)^i x_Y G(i) )) [see cayhamlem4 22863 (step 12.)]
    = U(0_Y ) [see cayhamlem1 22841 (step 11.)]
    = 0_A [see m2cpminv0 22736]

The transformations in steps 5., 6., 10., 12. and 13. are not mentioned in the proof provided in Wikipedia, since it makes no distinction between a matrix over a ring 𝑅 and its representation as matrix over the polynomial ring over the ring 𝑅 in general!

Theorem

cpmadurid 22842

The right-hand fundamental relation of the adjugate (see madurid 22619) applied to the characteristic matrix of a matrix. (Contributed by AV, 25-Oct-2019.)

⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝐵 = (Base‘𝐴)    &   ⊢ 𝐶 = (𝑁 CharPlyMat 𝑅)    &   ⊢ 𝑃 = (Poly₁‘𝑅)    &   ⊢ 𝑌 = (𝑁 Mat 𝑃)    &   ⊢ 𝑋 = (var₁‘𝑅)    &   ⊢ 𝑇 = (𝑁 matToPolyMat 𝑅)    &   ⊢ − = (-_g‘𝑌)    &   ⊢ · = ( ·_𝑠 ‘𝑌)    &   ⊢ 1 = (1_r‘𝑌)    &   ⊢ 𝐼 = ((𝑋 · 1 ) − (𝑇‘𝑀))    &   ⊢ 𝐽 = (𝑁 maAdju 𝑃)    &   ⊢ × = (._r‘𝑌)    ⇒   ⊢ ((𝑁 ∈ Fin ∧ 𝑅 ∈ CRing ∧ 𝑀 ∈ 𝐵) → (𝐼 × (𝐽‘𝐼)) = ((𝐶‘𝑀) · 1 ))

Theorem

cpmidgsum 22843*

Representation of the identity matrix multiplied with the characteristic polynomial of a matrix as group sum. (Contributed by AV, 7-Nov-2019.)

⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝐵 = (Base‘𝐴)    &   ⊢ 𝑃 = (Poly₁‘𝑅)    &   ⊢ 𝑌 = (𝑁 Mat 𝑃)    &   ⊢ 𝑋 = (var₁‘𝑅)    &   ⊢ ↑ = (._g‘(mulGrp‘𝑃))    &   ⊢ · = ( ·_𝑠 ‘𝑌)    &   ⊢ 1 = (1_r‘𝑌)    &   ⊢ 𝑈 = (algSc‘𝑃)    &   ⊢ 𝐶 = (𝑁 CharPlyMat 𝑅)    &   ⊢ 𝐾 = (𝐶‘𝑀)    &   ⊢ 𝐻 = (𝐾 · 1 )    ⇒   ⊢ ((𝑁 ∈ Fin ∧ 𝑅 ∈ CRing ∧ 𝑀 ∈ 𝐵) → 𝐻 = (𝑌 Σ_g (𝑛 ∈ ℕ₀ ↦ ((𝑛 ↑ 𝑋) · ((𝑈‘((coe₁‘𝐾)‘𝑛)) · 1 )))))

Theorem

cpmidgsumm2pm 22844*

Representation of the identity matrix multiplied with the characteristic polynomial of a matrix as group sum with a matrix to polynomial matrix transformation. (Contributed by AV, 13-Nov-2019.)

⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝐵 = (Base‘𝐴)    &   ⊢ 𝑃 = (Poly₁‘𝑅)    &   ⊢ 𝑌 = (𝑁 Mat 𝑃)    &   ⊢ 𝑋 = (var₁‘𝑅)    &   ⊢ ↑ = (._g‘(mulGrp‘𝑃))    &   ⊢ · = ( ·_𝑠 ‘𝑌)    &   ⊢ 1 = (1_r‘𝑌)    &   ⊢ 𝑈 = (algSc‘𝑃)    &   ⊢ 𝐶 = (𝑁 CharPlyMat 𝑅)    &   ⊢ 𝐾 = (𝐶‘𝑀)    &   ⊢ 𝐻 = (𝐾 · 1 )    &   ⊢ 𝑂 = (1_r‘𝐴)    &   ⊢ ∗ = ( ·_𝑠 ‘𝐴)    &   ⊢ 𝑇 = (𝑁 matToPolyMat 𝑅)    ⇒   ⊢ ((𝑁 ∈ Fin ∧ 𝑅 ∈ CRing ∧ 𝑀 ∈ 𝐵) → 𝐻 = (𝑌 Σ_g (𝑛 ∈ ℕ₀ ↦ ((𝑛 ↑ 𝑋) · (𝑇‘(((coe₁‘𝐾)‘𝑛) ∗ 𝑂))))))

Theorem

cpmidpmatlem1 22845*

Lemma 1 for cpmidpmat 22848. (Contributed by AV, 13-Nov-2019.)

⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝐵 = (Base‘𝐴)    &   ⊢ 𝑃 = (Poly₁‘𝑅)    &   ⊢ 𝑌 = (𝑁 Mat 𝑃)    &   ⊢ 𝑋 = (var₁‘𝑅)    &   ⊢ ↑ = (._g‘(mulGrp‘𝑃))    &   ⊢ · = ( ·_𝑠 ‘𝑌)    &   ⊢ 1 = (1_r‘𝑌)    &   ⊢ 𝑈 = (algSc‘𝑃)    &   ⊢ 𝐶 = (𝑁 CharPlyMat 𝑅)    &   ⊢ 𝐾 = (𝐶‘𝑀)    &   ⊢ 𝐻 = (𝐾 · 1 )    &   ⊢ 𝑂 = (1_r‘𝐴)    &   ⊢ ∗ = ( ·_𝑠 ‘𝐴)    &   ⊢ 𝑇 = (𝑁 matToPolyMat 𝑅)    &   ⊢ 𝐺 = (𝑘 ∈ ℕ₀ ↦ (((coe₁‘𝐾)‘𝑘) ∗ 𝑂))    ⇒   ⊢ (𝐿 ∈ ℕ₀ → (𝐺‘𝐿) = (((coe₁‘𝐾)‘𝐿) ∗ 𝑂))

Theorem

cpmidpmatlem2 22846*

Lemma 2 for cpmidpmat 22848. (Contributed by AV, 14-Nov-2019.) (Proof shortened by AV, 7-Dec-2019.)

⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝐵 = (Base‘𝐴)    &   ⊢ 𝑃 = (Poly₁‘𝑅)    &   ⊢ 𝑌 = (𝑁 Mat 𝑃)    &   ⊢ 𝑋 = (var₁‘𝑅)    &   ⊢ ↑ = (._g‘(mulGrp‘𝑃))    &   ⊢ · = ( ·_𝑠 ‘𝑌)    &   ⊢ 1 = (1_r‘𝑌)    &   ⊢ 𝑈 = (algSc‘𝑃)    &   ⊢ 𝐶 = (𝑁 CharPlyMat 𝑅)    &   ⊢ 𝐾 = (𝐶‘𝑀)    &   ⊢ 𝐻 = (𝐾 · 1 )    &   ⊢ 𝑂 = (1_r‘𝐴)    &   ⊢ ∗ = ( ·_𝑠 ‘𝐴)    &   ⊢ 𝑇 = (𝑁 matToPolyMat 𝑅)    &   ⊢ 𝐺 = (𝑘 ∈ ℕ₀ ↦ (((coe₁‘𝐾)‘𝑘) ∗ 𝑂))    ⇒   ⊢ ((𝑁 ∈ Fin ∧ 𝑅 ∈ CRing ∧ 𝑀 ∈ 𝐵) → 𝐺 ∈ (𝐵 ↑_m ℕ₀))

Theorem

cpmidpmatlem3 22847*

Lemma 3 for cpmidpmat 22848. (Contributed by AV, 14-Nov-2019.) (Proof shortened by AV, 7-Dec-2019.)

⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝐵 = (Base‘𝐴)    &   ⊢ 𝑃 = (Poly₁‘𝑅)    &   ⊢ 𝑌 = (𝑁 Mat 𝑃)    &   ⊢ 𝑋 = (var₁‘𝑅)    &   ⊢ ↑ = (._g‘(mulGrp‘𝑃))    &   ⊢ · = ( ·_𝑠 ‘𝑌)    &   ⊢ 1 = (1_r‘𝑌)    &   ⊢ 𝑈 = (algSc‘𝑃)    &   ⊢ 𝐶 = (𝑁 CharPlyMat 𝑅)    &   ⊢ 𝐾 = (𝐶‘𝑀)    &   ⊢ 𝐻 = (𝐾 · 1 )    &   ⊢ 𝑂 = (1_r‘𝐴)    &   ⊢ ∗ = ( ·_𝑠 ‘𝐴)    &   ⊢ 𝑇 = (𝑁 matToPolyMat 𝑅)    &   ⊢ 𝐺 = (𝑘 ∈ ℕ₀ ↦ (((coe₁‘𝐾)‘𝑘) ∗ 𝑂))    ⇒   ⊢ ((𝑁 ∈ Fin ∧ 𝑅 ∈ CRing ∧ 𝑀 ∈ 𝐵) → 𝐺 finSupp (0_g‘𝐴))

Theorem

cpmidpmat 22848*

Representation of the identity matrix multiplied with the characteristic polynomial of a matrix as polynomial over the ring of matrices. (Contributed by AV, 14-Nov-2019.) (Revised by AV, 7-Dec-2019.)

⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝐵 = (Base‘𝐴)    &   ⊢ 𝑃 = (Poly₁‘𝑅)    &   ⊢ 𝑌 = (𝑁 Mat 𝑃)    &   ⊢ 𝑋 = (var₁‘𝑅)    &   ⊢ ↑ = (._g‘(mulGrp‘𝑃))    &   ⊢ · = ( ·_𝑠 ‘𝑌)    &   ⊢ 1 = (1_r‘𝑌)    &   ⊢ 𝑈 = (algSc‘𝑃)    &   ⊢ 𝐶 = (𝑁 CharPlyMat 𝑅)    &   ⊢ 𝐾 = (𝐶‘𝑀)    &   ⊢ 𝐻 = (𝐾 · 1 )    &   ⊢ 𝑂 = (1_r‘𝐴)    &   ⊢ ∗ = ( ·_𝑠 ‘𝐴)    &   ⊢ 𝑇 = (𝑁 matToPolyMat 𝑅)    &   ⊢ 𝑊 = (Base‘𝑌)    &   ⊢ 𝑄 = (Poly₁‘𝐴)    &   ⊢ 𝑍 = (var₁‘𝐴)    &   ⊢ ∙ = ( ·_𝑠 ‘𝑄)    &   ⊢ 𝐸 = (._g‘(mulGrp‘𝑄))    &   ⊢ 𝐼 = (𝑁 pMatToMatPoly 𝑅)    ⇒   ⊢ ((𝑁 ∈ Fin ∧ 𝑅 ∈ CRing ∧ 𝑀 ∈ 𝐵) → (𝐼‘𝐻) = (𝑄 Σ_g (𝑛 ∈ ℕ₀ ↦ ((((coe₁‘𝐾)‘𝑛) ∗ 𝑂) ∙ (𝑛𝐸𝑍)))))

Theorem

cpmadugsumlemB 22849*

Lemma B for cpmadugsum 22853. (Contributed by AV, 2-Nov-2019.)

⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝐵 = (Base‘𝐴)    &   ⊢ 𝑃 = (Poly₁‘𝑅)    &   ⊢ 𝑌 = (𝑁 Mat 𝑃)    &   ⊢ 𝑇 = (𝑁 matToPolyMat 𝑅)    &   ⊢ 𝑋 = (var₁‘𝑅)    &   ⊢ ↑ = (._g‘(mulGrp‘𝑃))    &   ⊢ · = ( ·_𝑠 ‘𝑌)    &   ⊢ × = (._r‘𝑌)    &   ⊢ 1 = (1_r‘𝑌)    ⇒   ⊢ (((𝑁 ∈ Fin ∧ 𝑅 ∈ CRing ∧ 𝑀 ∈ 𝐵) ∧ (𝑠 ∈ ℕ₀ ∧ 𝑏 ∈ (𝐵 ↑_m (0...𝑠)))) → ((𝑋 · 1 ) × (𝑌 Σ_g (𝑖 ∈ (0...𝑠) ↦ ((𝑖 ↑ 𝑋) · (𝑇‘(𝑏‘𝑖)))))) = (𝑌 Σ_g (𝑖 ∈ (0...𝑠) ↦ (((𝑖 + 1) ↑ 𝑋) · (𝑇‘(𝑏‘𝑖))))))

Theorem

cpmadugsumlemC 22850*

Lemma C for cpmadugsum 22853. (Contributed by AV, 2-Nov-2019.)

⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝐵 = (Base‘𝐴)    &   ⊢ 𝑃 = (Poly₁‘𝑅)    &   ⊢ 𝑌 = (𝑁 Mat 𝑃)    &   ⊢ 𝑇 = (𝑁 matToPolyMat 𝑅)    &   ⊢ 𝑋 = (var₁‘𝑅)    &   ⊢ ↑ = (._g‘(mulGrp‘𝑃))    &   ⊢ · = ( ·_𝑠 ‘𝑌)    &   ⊢ × = (._r‘𝑌)    &   ⊢ 1 = (1_r‘𝑌)    ⇒   ⊢ (((𝑁 ∈ Fin ∧ 𝑅 ∈ CRing ∧ 𝑀 ∈ 𝐵) ∧ (𝑠 ∈ ℕ₀ ∧ 𝑏 ∈ (𝐵 ↑_m (0...𝑠)))) → ((𝑇‘𝑀) × (𝑌 Σ_g (𝑖 ∈ (0...𝑠) ↦ ((𝑖 ↑ 𝑋) · (𝑇‘(𝑏‘𝑖)))))) = (𝑌 Σ_g (𝑖 ∈ (0...𝑠) ↦ ((𝑖 ↑ 𝑋) · ((𝑇‘𝑀) × (𝑇‘(𝑏‘𝑖)))))))

Theorem

cpmadugsumlemF 22851*

Lemma F for cpmadugsum 22853. (Contributed by AV, 7-Nov-2019.)

⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝐵 = (Base‘𝐴)    &   ⊢ 𝑃 = (Poly₁‘𝑅)    &   ⊢ 𝑌 = (𝑁 Mat 𝑃)    &   ⊢ 𝑇 = (𝑁 matToPolyMat 𝑅)    &   ⊢ 𝑋 = (var₁‘𝑅)    &   ⊢ ↑ = (._g‘(mulGrp‘𝑃))    &   ⊢ · = ( ·_𝑠 ‘𝑌)    &   ⊢ × = (._r‘𝑌)    &   ⊢ 1 = (1_r‘𝑌)    &   ⊢ + = (+_g‘𝑌)    &   ⊢ − = (-_g‘𝑌)    ⇒   ⊢ (((𝑁 ∈ Fin ∧ 𝑅 ∈ CRing ∧ 𝑀 ∈ 𝐵) ∧ (𝑠 ∈ ℕ ∧ 𝑏 ∈ (𝐵 ↑_m (0...𝑠)))) → (((𝑋 · 1 ) × (𝑌 Σ_g (𝑖 ∈ (0...𝑠) ↦ ((𝑖 ↑ 𝑋) · (𝑇‘(𝑏‘𝑖)))))) − ((𝑇‘𝑀) × (𝑌 Σ_g (𝑖 ∈ (0...𝑠) ↦ ((𝑖 ↑ 𝑋) · (𝑇‘(𝑏‘𝑖))))))) = ((𝑌 Σ_g (𝑖 ∈ (1...𝑠) ↦ ((𝑖 ↑ 𝑋) · ((𝑇‘(𝑏‘(𝑖 − 1))) − ((𝑇‘𝑀) × (𝑇‘(𝑏‘𝑖))))))) + ((((𝑠 + 1) ↑ 𝑋) · (𝑇‘(𝑏‘𝑠))) − ((𝑇‘𝑀) × (𝑇‘(𝑏‘0))))))

Theorem

cpmadugsumfi 22852*

The product of the characteristic matrix of a given matrix and its adjunct represented as finite sum. (Contributed by AV, 7-Nov-2019.) (Proof shortened by AV, 29-Nov-2019.)

⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝐵 = (Base‘𝐴)    &   ⊢ 𝑃 = (Poly₁‘𝑅)    &   ⊢ 𝑌 = (𝑁 Mat 𝑃)    &   ⊢ 𝑇 = (𝑁 matToPolyMat 𝑅)    &   ⊢ 𝑋 = (var₁‘𝑅)    &   ⊢ ↑ = (._g‘(mulGrp‘𝑃))    &   ⊢ · = ( ·_𝑠 ‘𝑌)    &   ⊢ × = (._r‘𝑌)    &   ⊢ 1 = (1_r‘𝑌)    &   ⊢ + = (+_g‘𝑌)    &   ⊢ − = (-_g‘𝑌)    &   ⊢ 𝐼 = ((𝑋 · 1 ) − (𝑇‘𝑀))    &   ⊢ 𝐽 = (𝑁 maAdju 𝑃)    ⇒   ⊢ ((𝑁 ∈ Fin ∧ 𝑅 ∈ CRing ∧ 𝑀 ∈ 𝐵) → ∃𝑠 ∈ ℕ ∃𝑏 ∈ (𝐵 ↑_m (0...𝑠))(𝐼 × (𝐽‘𝐼)) = ((𝑌 Σ_g (𝑖 ∈ (1...𝑠) ↦ ((𝑖 ↑ 𝑋) · ((𝑇‘(𝑏‘(𝑖 − 1))) − ((𝑇‘𝑀) × (𝑇‘(𝑏‘𝑖))))))) + ((((𝑠 + 1) ↑ 𝑋) · (𝑇‘(𝑏‘𝑠))) − ((𝑇‘𝑀) × (𝑇‘(𝑏‘0))))))

Theorem

cpmadugsum 22853*

The product of the characteristic matrix of a given matrix and its adjunct represented as an infinite sum. (Contributed by AV, 10-Nov-2019.)

⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝐵 = (Base‘𝐴)    &   ⊢ 𝑃 = (Poly₁‘𝑅)    &   ⊢ 𝑌 = (𝑁 Mat 𝑃)    &   ⊢ 𝑇 = (𝑁 matToPolyMat 𝑅)    &   ⊢ 𝑋 = (var₁‘𝑅)    &   ⊢ ↑ = (._g‘(mulGrp‘𝑃))    &   ⊢ · = ( ·_𝑠 ‘𝑌)    &   ⊢ × = (._r‘𝑌)    &   ⊢ 1 = (1_r‘𝑌)    &   ⊢ + = (+_g‘𝑌)    &   ⊢ − = (-_g‘𝑌)    &   ⊢ 𝐼 = ((𝑋 · 1 ) − (𝑇‘𝑀))    &   ⊢ 𝐽 = (𝑁 maAdju 𝑃)    &   ⊢ 0 = (0_g‘𝑌)    &   ⊢ 𝐺 = (𝑛 ∈ ℕ₀ ↦ if(𝑛 = 0, ( 0 − ((𝑇‘𝑀) × (𝑇‘(𝑏‘0)))), if(𝑛 = (𝑠 + 1), (𝑇‘(𝑏‘𝑠)), if((𝑠 + 1) < 𝑛, 0 , ((𝑇‘(𝑏‘(𝑛 − 1))) − ((𝑇‘𝑀) × (𝑇‘(𝑏‘𝑛))))))))    ⇒   ⊢ ((𝑁 ∈ Fin ∧ 𝑅 ∈ CRing ∧ 𝑀 ∈ 𝐵) → ∃𝑠 ∈ ℕ ∃𝑏 ∈ (𝐵 ↑_m (0...𝑠))(𝐼 × (𝐽‘𝐼)) = (𝑌 Σ_g (𝑖 ∈ ℕ₀ ↦ ((𝑖 ↑ 𝑋) · (𝐺‘𝑖)))))

Theorem

cpmidgsum2 22854*

Representation of the identity matrix multiplied with the characteristic polynomial of a matrix as another group sum. (Contributed by AV, 10-Nov-2019.)

⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝐵 = (Base‘𝐴)    &   ⊢ 𝑃 = (Poly₁‘𝑅)    &   ⊢ 𝑌 = (𝑁 Mat 𝑃)    &   ⊢ 𝑇 = (𝑁 matToPolyMat 𝑅)    &   ⊢ 𝑋 = (var₁‘𝑅)    &   ⊢ ↑ = (._g‘(mulGrp‘𝑃))    &   ⊢ · = ( ·_𝑠 ‘𝑌)    &   ⊢ × = (._r‘𝑌)    &   ⊢ 1 = (1_r‘𝑌)    &   ⊢ + = (+_g‘𝑌)    &   ⊢ − = (-_g‘𝑌)    &   ⊢ 𝐼 = ((𝑋 · 1 ) − (𝑇‘𝑀))    &   ⊢ 𝐽 = (𝑁 maAdju 𝑃)    &   ⊢ 0 = (0_g‘𝑌)    &   ⊢ 𝐺 = (𝑛 ∈ ℕ₀ ↦ if(𝑛 = 0, ( 0 − ((𝑇‘𝑀) × (𝑇‘(𝑏‘0)))), if(𝑛 = (𝑠 + 1), (𝑇‘(𝑏‘𝑠)), if((𝑠 + 1) < 𝑛, 0 , ((𝑇‘(𝑏‘(𝑛 − 1))) − ((𝑇‘𝑀) × (𝑇‘(𝑏‘𝑛))))))))    &   ⊢ 𝐶 = (𝑁 CharPlyMat 𝑅)    &   ⊢ 𝐾 = (𝐶‘𝑀)    &   ⊢ 𝐻 = (𝐾 · 1 )    ⇒   ⊢ ((𝑁 ∈ Fin ∧ 𝑅 ∈ CRing ∧ 𝑀 ∈ 𝐵) → ∃𝑠 ∈ ℕ ∃𝑏 ∈ (𝐵 ↑_m (0...𝑠))𝐻 = (𝑌 Σ_g (𝑖 ∈ ℕ₀ ↦ ((𝑖 ↑ 𝑋) · (𝐺‘𝑖)))))

Theorem

cpmidg2sum 22855*

Equality of two sums representing the identity matrix multiplied with the characteristic polynomial of a matrix. (Contributed by AV, 11-Nov-2019.)

⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝐵 = (Base‘𝐴)    &   ⊢ 𝑃 = (Poly₁‘𝑅)    &   ⊢ 𝑌 = (𝑁 Mat 𝑃)    &   ⊢ 𝑇 = (𝑁 matToPolyMat 𝑅)    &   ⊢ 𝑋 = (var₁‘𝑅)    &   ⊢ ↑ = (._g‘(mulGrp‘𝑃))    &   ⊢ · = ( ·_𝑠 ‘𝑌)    &   ⊢ × = (._r‘𝑌)    &   ⊢ 1 = (1_r‘𝑌)    &   ⊢ + = (+_g‘𝑌)    &   ⊢ − = (-_g‘𝑌)    &   ⊢ 𝐼 = ((𝑋 · 1 ) − (𝑇‘𝑀))    &   ⊢ 𝐽 = (𝑁 maAdju 𝑃)    &   ⊢ 0 = (0_g‘𝑌)    &   ⊢ 𝐺 = (𝑛 ∈ ℕ₀ ↦ if(𝑛 = 0, ( 0 − ((𝑇‘𝑀) × (𝑇‘(𝑏‘0)))), if(𝑛 = (𝑠 + 1), (𝑇‘(𝑏‘𝑠)), if((𝑠 + 1) < 𝑛, 0 , ((𝑇‘(𝑏‘(𝑛 − 1))) − ((𝑇‘𝑀) × (𝑇‘(𝑏‘𝑛))))))))    &   ⊢ 𝐶 = (𝑁 CharPlyMat 𝑅)    &   ⊢ 𝐾 = (𝐶‘𝑀)    &   ⊢ 𝑈 = (algSc‘𝑃)    ⇒   ⊢ ((𝑁 ∈ Fin ∧ 𝑅 ∈ CRing ∧ 𝑀 ∈ 𝐵) → ∃𝑠 ∈ ℕ ∃𝑏 ∈ (𝐵 ↑_m (0...𝑠))(𝑌 Σ_g (𝑖 ∈ ℕ₀ ↦ ((𝑖 ↑ 𝑋) · ((𝑈‘((coe₁‘𝐾)‘𝑖)) · 1 )))) = (𝑌 Σ_g (𝑖 ∈ ℕ₀ ↦ ((𝑖 ↑ 𝑋) · (𝐺‘𝑖)))))

Theorem

cpmadumatpolylem1 22856*

Lemma 1 for cpmadumatpoly 22858. (Contributed by AV, 20-Nov-2019.) (Revised by AV, 15-Dec-2019.)

⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝐵 = (Base‘𝐴)    &   ⊢ 𝑃 = (Poly₁‘𝑅)    &   ⊢ 𝑌 = (𝑁 Mat 𝑃)    &   ⊢ 𝑇 = (𝑁 matToPolyMat 𝑅)    &   ⊢ × = (._r‘𝑌)    &   ⊢ − = (-_g‘𝑌)    &   ⊢ 0 = (0_g‘𝑌)    &   ⊢ 𝐺 = (𝑛 ∈ ℕ₀ ↦ if(𝑛 = 0, ( 0 − ((𝑇‘𝑀) × (𝑇‘(𝑏‘0)))), if(𝑛 = (𝑠 + 1), (𝑇‘(𝑏‘𝑠)), if((𝑠 + 1) < 𝑛, 0 , ((𝑇‘(𝑏‘(𝑛 − 1))) − ((𝑇‘𝑀) × (𝑇‘(𝑏‘𝑛))))))))    &   ⊢ 𝑆 = (𝑁 ConstPolyMat 𝑅)    &   ⊢ · = ( ·_𝑠 ‘𝑌)    &   ⊢ 1 = (1_r‘𝑌)    &   ⊢ 𝑍 = (var₁‘𝑅)    &   ⊢ 𝐷 = ((𝑍 · 1 ) − (𝑇‘𝑀))    &   ⊢ 𝐽 = (𝑁 maAdju 𝑃)    &   ⊢ 𝑊 = (Base‘𝑌)    &   ⊢ 𝑄 = (Poly₁‘𝐴)    &   ⊢ 𝑋 = (var₁‘𝐴)    &   ⊢ ∗ = ( ·_𝑠 ‘𝑄)    &   ⊢ ↑ = (._g‘(mulGrp‘𝑄))    &   ⊢ 𝑈 = (𝑁 cPolyMatToMat 𝑅)    ⇒   ⊢ ((((𝑁 ∈ Fin ∧ 𝑅 ∈ CRing ∧ 𝑀 ∈ 𝐵) ∧ 𝑠 ∈ ℕ) ∧ 𝑏 ∈ (𝐵 ↑_m (0...𝑠))) → (𝑈 ∘ 𝐺) ∈ (𝐵 ↑_m ℕ₀))

Theorem

cpmadumatpolylem2 22857*

Lemma 2 for cpmadumatpoly 22858. (Contributed by AV, 20-Nov-2019.) (Revised by AV, 15-Dec-2019.)

⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝐵 = (Base‘𝐴)    &   ⊢ 𝑃 = (Poly₁‘𝑅)    &   ⊢ 𝑌 = (𝑁 Mat 𝑃)    &   ⊢ 𝑇 = (𝑁 matToPolyMat 𝑅)    &   ⊢ × = (._r‘𝑌)    &   ⊢ − = (-_g‘𝑌)    &   ⊢ 0 = (0_g‘𝑌)    &   ⊢ 𝐺 = (𝑛 ∈ ℕ₀ ↦ if(𝑛 = 0, ( 0 − ((𝑇‘𝑀) × (𝑇‘(𝑏‘0)))), if(𝑛 = (𝑠 + 1), (𝑇‘(𝑏‘𝑠)), if((𝑠 + 1) < 𝑛, 0 , ((𝑇‘(𝑏‘(𝑛 − 1))) − ((𝑇‘𝑀) × (𝑇‘(𝑏‘𝑛))))))))    &   ⊢ 𝑆 = (𝑁 ConstPolyMat 𝑅)    &   ⊢ · = ( ·_𝑠 ‘𝑌)    &   ⊢ 1 = (1_r‘𝑌)    &   ⊢ 𝑍 = (var₁‘𝑅)    &   ⊢ 𝐷 = ((𝑍 · 1 ) − (𝑇‘𝑀))    &   ⊢ 𝐽 = (𝑁 maAdju 𝑃)    &   ⊢ 𝑊 = (Base‘𝑌)    &   ⊢ 𝑄 = (Poly₁‘𝐴)    &   ⊢ 𝑋 = (var₁‘𝐴)    &   ⊢ ∗ = ( ·_𝑠 ‘𝑄)    &   ⊢ ↑ = (._g‘(mulGrp‘𝑄))    &   ⊢ 𝑈 = (𝑁 cPolyMatToMat 𝑅)    ⇒   ⊢ ((((𝑁 ∈ Fin ∧ 𝑅 ∈ CRing ∧ 𝑀 ∈ 𝐵) ∧ 𝑠 ∈ ℕ) ∧ 𝑏 ∈ (𝐵 ↑_m (0...𝑠))) → (𝑈 ∘ 𝐺) finSupp (0_g‘𝐴))

Theorem

cpmadumatpoly 22858*

The product of the characteristic matrix of a given matrix and its adjunct represented as a polynomial over matrices. (Contributed by AV, 20-Nov-2019.) (Revised by AV, 7-Dec-2019.)

⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝐵 = (Base‘𝐴)    &   ⊢ 𝑃 = (Poly₁‘𝑅)    &   ⊢ 𝑌 = (𝑁 Mat 𝑃)    &   ⊢ 𝑇 = (𝑁 matToPolyMat 𝑅)    &   ⊢ × = (._r‘𝑌)    &   ⊢ − = (-_g‘𝑌)    &   ⊢ 0 = (0_g‘𝑌)    &   ⊢ 𝐺 = (𝑛 ∈ ℕ₀ ↦ if(𝑛 = 0, ( 0 − ((𝑇‘𝑀) × (𝑇‘(𝑏‘0)))), if(𝑛 = (𝑠 + 1), (𝑇‘(𝑏‘𝑠)), if((𝑠 + 1) < 𝑛, 0 , ((𝑇‘(𝑏‘(𝑛 − 1))) − ((𝑇‘𝑀) × (𝑇‘(𝑏‘𝑛))))))))    &   ⊢ 𝑆 = (𝑁 ConstPolyMat 𝑅)    &   ⊢ · = ( ·_𝑠 ‘𝑌)    &   ⊢ 1 = (1_r‘𝑌)    &   ⊢ 𝑍 = (var₁‘𝑅)    &   ⊢ 𝐷 = ((𝑍 · 1 ) − (𝑇‘𝑀))    &   ⊢ 𝐽 = (𝑁 maAdju 𝑃)    &   ⊢ 𝑊 = (Base‘𝑌)    &   ⊢ 𝑄 = (Poly₁‘𝐴)    &   ⊢ 𝑋 = (var₁‘𝐴)    &   ⊢ ∗ = ( ·_𝑠 ‘𝑄)    &   ⊢ ↑ = (._g‘(mulGrp‘𝑄))    &   ⊢ 𝑈 = (𝑁 cPolyMatToMat 𝑅)    &   ⊢ 𝐼 = (𝑁 pMatToMatPoly 𝑅)    ⇒   ⊢ ((𝑁 ∈ Fin ∧ 𝑅 ∈ CRing ∧ 𝑀 ∈ 𝐵) → ∃𝑠 ∈ ℕ ∃𝑏 ∈ (𝐵 ↑_m (0...𝑠))(𝐼‘(𝐷 × (𝐽‘𝐷))) = (𝑄 Σ_g (𝑛 ∈ ℕ₀ ↦ ((𝑈‘(𝐺‘𝑛)) ∗ (𝑛 ↑ 𝑋)))))

Theorem

cayhamlem2 22859

Lemma for cayhamlem3 22862. (Contributed by AV, 24-Nov-2019.)

⊢ 𝐾 = (Base‘𝑅)    &   ⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝐵 = (Base‘𝐴)    &   ⊢ 1 = (1_r‘𝐴)    &   ⊢ ∗ = ( ·_𝑠 ‘𝐴)    &   ⊢ ↑ = (._g‘(mulGrp‘𝐴))    &   ⊢ · = (._r‘𝐴)    ⇒   ⊢ (((𝑁 ∈ Fin ∧ 𝑅 ∈ CRing ∧ 𝑀 ∈ 𝐵) ∧ (𝐻 ∈ (𝐾 ↑_m ℕ₀) ∧ 𝐿 ∈ ℕ₀)) → ((𝐻‘𝐿) ∗ (𝐿 ↑ 𝑀)) = ((𝐿 ↑ 𝑀) · ((𝐻‘𝐿) ∗ 1 )))

Theorem

chcoeffeqlem 22860*

Lemma for chcoeffeq 22861. (Contributed by AV, 21-Nov-2019.) (Proof shortened by AV, 7-Dec-2019.) (Revised by AV, 15-Dec-2019.)

⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝐵 = (Base‘𝐴)    &   ⊢ 𝑃 = (Poly₁‘𝑅)    &   ⊢ 𝑌 = (𝑁 Mat 𝑃)    &   ⊢ × = (._r‘𝑌)    &   ⊢ − = (-_g‘𝑌)    &   ⊢ 0 = (0_g‘𝑌)    &   ⊢ 𝑇 = (𝑁 matToPolyMat 𝑅)    &   ⊢ 𝐶 = (𝑁 CharPlyMat 𝑅)    &   ⊢ 𝐾 = (𝐶‘𝑀)    &   ⊢ 𝐺 = (𝑛 ∈ ℕ₀ ↦ if(𝑛 = 0, ( 0 − ((𝑇‘𝑀) × (𝑇‘(𝑏‘0)))), if(𝑛 = (𝑠 + 1), (𝑇‘(𝑏‘𝑠)), if((𝑠 + 1) < 𝑛, 0 , ((𝑇‘(𝑏‘(𝑛 − 1))) − ((𝑇‘𝑀) × (𝑇‘(𝑏‘𝑛))))))))    &   ⊢ 𝑊 = (Base‘𝑌)    &   ⊢ 1 = (1_r‘𝐴)    &   ⊢ ∗ = ( ·_𝑠 ‘𝐴)    &   ⊢ 𝑈 = (𝑁 cPolyMatToMat 𝑅)    ⇒   ⊢ (((𝑁 ∈ Fin ∧ 𝑅 ∈ CRing ∧ 𝑀 ∈ 𝐵) ∧ (𝑠 ∈ ℕ ∧ 𝑏 ∈ (𝐵 ↑_m (0...𝑠)))) → (((Poly₁‘𝐴) Σ_g (𝑛 ∈ ℕ₀ ↦ ((𝑈‘(𝐺‘𝑛))( ·_𝑠 ‘(Poly₁‘𝐴))(𝑛(._g‘(mulGrp‘(Poly₁‘𝐴)))(var₁‘𝐴))))) = ((Poly₁‘𝐴) Σ_g (𝑛 ∈ ℕ₀ ↦ ((((coe₁‘𝐾)‘𝑛) ∗ 1 )( ·_𝑠 ‘(Poly₁‘𝐴))(𝑛(._g‘(mulGrp‘(Poly₁‘𝐴)))(var₁‘𝐴))))) → ∀𝑛 ∈ ℕ₀ (𝑈‘(𝐺‘𝑛)) = (((coe₁‘𝐾)‘𝑛) ∗ 1 )))

Theorem

chcoeffeq 22861*

The coefficients of the characteristic polynomial multiplied with the identity matrix represented by (transformed) ring elements obtained from the adjunct of the characteristic matrix. (Contributed by AV, 21-Nov-2019.) (Proof shortened by AV, 8-Dec-2019.) (Revised by AV, 15-Dec-2019.)

⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝐵 = (Base‘𝐴)    &   ⊢ 𝑃 = (Poly₁‘𝑅)    &   ⊢ 𝑌 = (𝑁 Mat 𝑃)    &   ⊢ × = (._r‘𝑌)    &   ⊢ − = (-_g‘𝑌)    &   ⊢ 0 = (0_g‘𝑌)    &   ⊢ 𝑇 = (𝑁 matToPolyMat 𝑅)    &   ⊢ 𝐶 = (𝑁 CharPlyMat 𝑅)    &   ⊢ 𝐾 = (𝐶‘𝑀)    &   ⊢ 𝐺 = (𝑛 ∈ ℕ₀ ↦ if(𝑛 = 0, ( 0 − ((𝑇‘𝑀) × (𝑇‘(𝑏‘0)))), if(𝑛 = (𝑠 + 1), (𝑇‘(𝑏‘𝑠)), if((𝑠 + 1) < 𝑛, 0 , ((𝑇‘(𝑏‘(𝑛 − 1))) − ((𝑇‘𝑀) × (𝑇‘(𝑏‘𝑛))))))))    &   ⊢ 𝑊 = (Base‘𝑌)    &   ⊢ 1 = (1_r‘𝐴)    &   ⊢ ∗ = ( ·_𝑠 ‘𝐴)    &   ⊢ 𝑈 = (𝑁 cPolyMatToMat 𝑅)    ⇒   ⊢ ((𝑁 ∈ Fin ∧ 𝑅 ∈ CRing ∧ 𝑀 ∈ 𝐵) → ∃𝑠 ∈ ℕ ∃𝑏 ∈ (𝐵 ↑_m (0...𝑠))∀𝑛 ∈ ℕ₀ (𝑈‘(𝐺‘𝑛)) = (((coe₁‘𝐾)‘𝑛) ∗ 1 ))

Theorem

cayhamlem3 22862*

Lemma for cayhamlem4 22863. (Contributed by AV, 24-Nov-2019.) (Revised by AV, 15-Dec-2019.)

⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝐵 = (Base‘𝐴)    &   ⊢ 𝑃 = (Poly₁‘𝑅)    &   ⊢ 𝑌 = (𝑁 Mat 𝑃)    &   ⊢ × = (._r‘𝑌)    &   ⊢ − = (-_g‘𝑌)    &   ⊢ 0 = (0_g‘𝑌)    &   ⊢ 𝑇 = (𝑁 matToPolyMat 𝑅)    &   ⊢ 𝐶 = (𝑁 CharPlyMat 𝑅)    &   ⊢ 𝐾 = (𝐶‘𝑀)    &   ⊢ 𝐺 = (𝑛 ∈ ℕ₀ ↦ if(𝑛 = 0, ( 0 − ((𝑇‘𝑀) × (𝑇‘(𝑏‘0)))), if(𝑛 = (𝑠 + 1), (𝑇‘(𝑏‘𝑠)), if((𝑠 + 1) < 𝑛, 0 , ((𝑇‘(𝑏‘(𝑛 − 1))) − ((𝑇‘𝑀) × (𝑇‘(𝑏‘𝑛))))))))    &   ⊢ 𝑊 = (Base‘𝑌)    &   ⊢ 1 = (1_r‘𝐴)    &   ⊢ ∗ = ( ·_𝑠 ‘𝐴)    &   ⊢ 𝑈 = (𝑁 cPolyMatToMat 𝑅)    &   ⊢ ↑ = (._g‘(mulGrp‘𝐴))    &   ⊢ · = (._r‘𝐴)    ⇒   ⊢ ((𝑁 ∈ Fin ∧ 𝑅 ∈ CRing ∧ 𝑀 ∈ 𝐵) → ∃𝑠 ∈ ℕ ∃𝑏 ∈ (𝐵 ↑_m (0...𝑠))(𝐴 Σ_g (𝑛 ∈ ℕ₀ ↦ (((coe₁‘𝐾)‘𝑛) ∗ (𝑛 ↑ 𝑀)))) = (𝐴 Σ_g (𝑛 ∈ ℕ₀ ↦ ((𝑛 ↑ 𝑀) · (𝑈‘(𝐺‘𝑛))))))

Theorem

cayhamlem4 22863*

Lemma for cayleyhamilton 22865. (Contributed by AV, 25-Nov-2019.) (Revised by AV, 15-Dec-2019.)

⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝐵 = (Base‘𝐴)    &   ⊢ 𝑃 = (Poly₁‘𝑅)    &   ⊢ 𝑌 = (𝑁 Mat 𝑃)    &   ⊢ × = (._r‘𝑌)    &   ⊢ − = (-_g‘𝑌)    &   ⊢ 0 = (0_g‘𝑌)    &   ⊢ 𝑇 = (𝑁 matToPolyMat 𝑅)    &   ⊢ 𝐶 = (𝑁 CharPlyMat 𝑅)    &   ⊢ 𝐾 = (𝐶‘𝑀)    &   ⊢ 𝐺 = (𝑛 ∈ ℕ₀ ↦ if(𝑛 = 0, ( 0 − ((𝑇‘𝑀) × (𝑇‘(𝑏‘0)))), if(𝑛 = (𝑠 + 1), (𝑇‘(𝑏‘𝑠)), if((𝑠 + 1) < 𝑛, 0 , ((𝑇‘(𝑏‘(𝑛 − 1))) − ((𝑇‘𝑀) × (𝑇‘(𝑏‘𝑛))))))))    &   ⊢ 𝑊 = (Base‘𝑌)    &   ⊢ 1 = (1_r‘𝐴)    &   ⊢ ∗ = ( ·_𝑠 ‘𝐴)    &   ⊢ 𝑈 = (𝑁 cPolyMatToMat 𝑅)    &   ⊢ ↑ = (._g‘(mulGrp‘𝐴))    &   ⊢ 𝐸 = (._g‘(mulGrp‘𝑌))    ⇒   ⊢ ((𝑁 ∈ Fin ∧ 𝑅 ∈ CRing ∧ 𝑀 ∈ 𝐵) → ∃𝑠 ∈ ℕ ∃𝑏 ∈ (𝐵 ↑_m (0...𝑠))(𝐴 Σ_g (𝑛 ∈ ℕ₀ ↦ (((coe₁‘𝐾)‘𝑛) ∗ (𝑛 ↑ 𝑀)))) = (𝑈‘(𝑌 Σ_g (𝑛 ∈ ℕ₀ ↦ ((𝑛𝐸(𝑇‘𝑀)) × (𝐺‘𝑛))))))

Theorem

cayleyhamilton0 22864*

The Cayley-Hamilton theorem: A matrix over a commutative ring "satisfies its own characteristic equation". This version of cayleyhamilton 22865 provides definitions not used in the theorem itself, but in its proof to make it clearer, more readable and shorter compared with a proof without them (see cayleyhamiltonALT 22866)! (Contributed by AV, 25-Nov-2019.) (Revised by AV, 15-Dec-2019.)

⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝐵 = (Base‘𝐴)    &   ⊢ 0 = (0_g‘𝐴)    &   ⊢ 1 = (1_r‘𝐴)    &   ⊢ ∗ = ( ·_𝑠 ‘𝐴)    &   ⊢ ↑ = (._g‘(mulGrp‘𝐴))    &   ⊢ 𝐶 = (𝑁 CharPlyMat 𝑅)    &   ⊢ 𝐾 = (coe₁‘(𝐶‘𝑀))    &   ⊢ 𝑃 = (Poly₁‘𝑅)    &   ⊢ 𝑌 = (𝑁 Mat 𝑃)    &   ⊢ × = (._r‘𝑌)    &   ⊢ − = (-_g‘𝑌)    &   ⊢ 𝑍 = (0_g‘𝑌)    &   ⊢ 𝑊 = (Base‘𝑌)    &   ⊢ 𝐸 = (._g‘(mulGrp‘𝑌))    &   ⊢ 𝑇 = (𝑁 matToPolyMat 𝑅)    &   ⊢ 𝐺 = (𝑛 ∈ ℕ₀ ↦ if(𝑛 = 0, (𝑍 − ((𝑇‘𝑀) × (𝑇‘(𝑏‘0)))), if(𝑛 = (𝑠 + 1), (𝑇‘(𝑏‘𝑠)), if((𝑠 + 1) < 𝑛, 𝑍, ((𝑇‘(𝑏‘(𝑛 − 1))) − ((𝑇‘𝑀) × (𝑇‘(𝑏‘𝑛))))))))    &   ⊢ 𝑈 = (𝑁 cPolyMatToMat 𝑅)    ⇒   ⊢ ((𝑁 ∈ Fin ∧ 𝑅 ∈ CRing ∧ 𝑀 ∈ 𝐵) → (𝐴 Σ_g (𝑛 ∈ ℕ₀ ↦ ((𝐾‘𝑛) ∗ (𝑛 ↑ 𝑀)))) = 0 )

Theorem

cayleyhamilton 22865*

The Cayley-Hamilton theorem: A matrix over a commutative ring "satisfies its own characteristic equation", see theorem 7.8 in [Roman] p. 170 (without proof!), or theorem 3.1 in [Lang] p. 561. In other words, a matrix over a commutative ring "inserted" into its characteristic polynomial results in zero. This is Metamath 100 proof #49. (Contributed by Alexander van der Vekens, 25-Nov-2019.)

⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝐵 = (Base‘𝐴)    &   ⊢ 0 = (0_g‘𝐴)    &   ⊢ 𝐶 = (𝑁 CharPlyMat 𝑅)    &   ⊢ 𝐾 = (coe₁‘(𝐶‘𝑀))    &   ⊢ ∗ = ( ·_𝑠 ‘𝐴)    &   ⊢ ↑ = (._g‘(mulGrp‘𝐴))    ⇒   ⊢ ((𝑁 ∈ Fin ∧ 𝑅 ∈ CRing ∧ 𝑀 ∈ 𝐵) → (𝐴 Σ_g (𝑛 ∈ ℕ₀ ↦ ((𝐾‘𝑛) ∗ (𝑛 ↑ 𝑀)))) = 0 )

Theorem

cayleyhamiltonALT 22866*

Alternate proof of cayleyhamilton 22865, the Cayley-Hamilton theorem. This proof does not use cayleyhamilton0 22864 directly, but has the same structure as the proof of cayleyhamilton0 22864. In contrast to the proof of cayleyhamilton0 22864, only the definitions required to formulate the theorem itself are used, causing the definitions used in the lemmas being expanded, which makes the proof longer and more difficult to read. (Contributed by AV, 25-Nov-2019.) (New usage is discouraged.) (Proof modification is discouraged.)

⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝐵 = (Base‘𝐴)    &   ⊢ 0 = (0_g‘𝐴)    &   ⊢ 𝐶 = (𝑁 CharPlyMat 𝑅)    &   ⊢ 𝐾 = (coe₁‘(𝐶‘𝑀))    &   ⊢ ∗ = ( ·_𝑠 ‘𝐴)    &   ⊢ ↑ = (._g‘(mulGrp‘𝐴))    ⇒   ⊢ ((𝑁 ∈ Fin ∧ 𝑅 ∈ CRing ∧ 𝑀 ∈ 𝐵) → (𝐴 Σ_g (𝑛 ∈ ℕ₀ ↦ ((𝐾‘𝑛) ∗ (𝑛 ↑ 𝑀)))) = 0 )

Theorem

cayleyhamilton1 22867*

The Cayley-Hamilton theorem: A matrix over a commutative ring "satisfies its own characteristic equation", or, in other words, a matrix over a commutative ring "inserted" into its characteristic polynomial results in zero. In this variant of cayleyhamilton 22865, the meaning of "inserted" is made more transparent: If the characteristic polynomial is a polynomial with coefficients (𝐹‘𝑛), then a matrix over a commutative ring "inserted" into its characteristic polynomial is the sum of these coefficients multiplied with the corresponding power of the matrix. (Contributed by AV, 25-Nov-2019.)

⊢ 𝐴 = (𝑁 Mat 𝑅)    &   ⊢ 𝐵 = (Base‘𝐴)    &   ⊢ 0 = (0_g‘𝐴)    &   ⊢ 𝐶 = (𝑁 CharPlyMat 𝑅)    &   ⊢ 𝐾 = (coe₁‘(𝐶‘𝑀))    &   ⊢ ∗ = ( ·_𝑠 ‘𝐴)    &   ⊢ ↑ = (._g‘(mulGrp‘𝐴))    &   ⊢ 𝐿 = (Base‘𝑅)    &   ⊢ 𝑋 = (var₁‘𝑅)    &   ⊢ 𝑃 = (Poly₁‘𝑅)    &   ⊢ · = ( ·_𝑠 ‘𝑃)    &   ⊢ 𝐸 = (._g‘(mulGrp‘𝑃))    &   ⊢ 𝑍 = (0_g‘𝑅)    ⇒   ⊢ (((𝑁 ∈ Fin ∧ 𝑅 ∈ CRing ∧ 𝑀 ∈ 𝐵) ∧ (𝐹 ∈ (𝐿 ↑_m ℕ₀) ∧ 𝐹 finSupp 𝑍)) → ((𝐶‘𝑀) = (𝑃 Σ_g (𝑛 ∈ ℕ₀ ↦ ((𝐹‘𝑛) · (𝑛𝐸𝑋)))) → (𝐴 Σ_g (𝑛 ∈ ℕ₀ ↦ ((𝐹‘𝑛) ∗ (𝑛 ↑ 𝑀)))) = 0 ))

PART 12  BASIC TOPOLOGY

12.1  Topology

12.1.1  Topological spaces

A topology on a set is a set of subsets of that set, called open sets, which satisfy certain conditions. One condition is that the whole set be an open set. Therefore, a set is recoverable from a topology on it (as its union, see toponuni 22889), and it may sometimes be more convenient to consider topologies without reference to the underlying set. This is why we define successively the class of topologies (df-top 22869), then the function which associates with a set the set of topologies on it (df-topon 22886), and finally the class of topological spaces, as extensible structures having an underlying set and a topology on it (df-topsp 22908). Of course, a topology is the same thing as a topology on a set (see toprntopon 22900).

12.1.1.1  Topologies

Syntax

ctop 22868

Syntax for the class of topologies.

class Top

Definition

df-top 22869*

Define the class of topologies. It is a proper class (see topnex 22971). See istopg 22870 and istop2g 22871 for the corresponding characterizations, using respectively binary intersections like in this definition and nonempty finite intersections.

The final form of the definition is due to Bourbaki (Def. 1 of [BourbakiTop1] p. I.1), while the idea of defining a topology in terms of its open sets is due to Aleksandrov. For the convoluted history of the definitions of these notions, see

Gregory H. Moore, The emergence of open sets, closed sets, and limit points in analysis and topology, Historia Mathematica 35 (2008) 220--241.

(Contributed by NM, 3-Mar-2006.) (Revised by BJ, 20-Oct-2018.)

⊢ Top = {𝑥 ∣ (∀𝑦 ∈ 𝒫 𝑥∪ 𝑦 ∈ 𝑥 ∧ ∀𝑦 ∈ 𝑥 ∀𝑧 ∈ 𝑥 (𝑦 ∩ 𝑧) ∈ 𝑥)}

Theorem

istopg 22870*

Express the predicate "𝐽 is a topology". See istop2g 22871 for another characterization using nonempty finite intersections instead of binary intersections.

Note: In the literature, a topology is often represented by a calligraphic letter T, which resembles the letter J. This confusion may have led to J being used by some authors (e.g., K. D. Joshi, Introduction to General Topology (1983), p. 114) and it is convenient for us since we later use 𝑇 to represent linear transformations (operators). (Contributed by Stefan Allan, 3-Mar-2006.) (Revised by Mario Carneiro, 11-Nov-2013.)

⊢ (𝐽 ∈ 𝐴 → (𝐽 ∈ Top ↔ (∀𝑥(𝑥 ⊆ 𝐽 → ∪ 𝑥 ∈ 𝐽) ∧ ∀𝑥 ∈ 𝐽 ∀𝑦 ∈ 𝐽 (𝑥 ∩ 𝑦) ∈ 𝐽)))

Theorem

istop2g 22871*

Express the predicate "𝐽 is a topology" using nonempty finite intersections instead of binary intersections as in istopg 22870. (Contributed by NM, 19-Jul-2006.)

⊢ (𝐽 ∈ 𝐴 → (𝐽 ∈ Top ↔ (∀𝑥(𝑥 ⊆ 𝐽 → ∪ 𝑥 ∈ 𝐽) ∧ ∀𝑥((𝑥 ⊆ 𝐽 ∧ 𝑥 ≠ ∅ ∧ 𝑥 ∈ Fin) → ∩ 𝑥 ∈ 𝐽))))

Theorem

uniopn 22872

The union of a subset of a topology (that is, the union of any family of open sets of a topology) is an open set. (Contributed by Stefan Allan, 27-Feb-2006.)

⊢ ((𝐽 ∈ Top ∧ 𝐴 ⊆ 𝐽) → ∪ 𝐴 ∈ 𝐽)

Theorem

iunopn 22873*

The indexed union of a subset of a topology is an open set. (Contributed by NM, 5-Oct-2006.)

⊢ ((𝐽 ∈ Top ∧ ∀𝑥 ∈ 𝐴 𝐵 ∈ 𝐽) → ∪ 𝑥 ∈ 𝐴 𝐵 ∈ 𝐽)

Theorem

inopn 22874

The intersection of two open sets of a topology is an open set. (Contributed by NM, 17-Jul-2006.)

⊢ ((𝐽 ∈ Top ∧ 𝐴 ∈ 𝐽 ∧ 𝐵 ∈ 𝐽) → (𝐴 ∩ 𝐵) ∈ 𝐽)

Theorem

fitop 22875

A topology is closed under finite intersections. (Contributed by Jeff Hankins, 7-Oct-2009.)

⊢ (𝐽 ∈ Top → (fi‘𝐽) = 𝐽)

Theorem

fiinopn 22876

The intersection of a nonempty finite family of open sets is open. (Contributed by FL, 20-Apr-2012.)

⊢ (𝐽 ∈ Top → ((𝐴 ⊆ 𝐽 ∧ 𝐴 ≠ ∅ ∧ 𝐴 ∈ Fin) → ∩ 𝐴 ∈ 𝐽))

Theorem

iinopn 22877*

The intersection of a nonempty finite family of open sets is open. (Contributed by Mario Carneiro, 14-Sep-2014.)

⊢ ((𝐽 ∈ Top ∧ (𝐴 ∈ Fin ∧ 𝐴 ≠ ∅ ∧ ∀𝑥 ∈ 𝐴 𝐵 ∈ 𝐽)) → ∩ 𝑥 ∈ 𝐴 𝐵 ∈ 𝐽)

Theorem

unopn 22878

The union of two open sets is open. (Contributed by Jeff Madsen, 2-Sep-2009.)

⊢ ((𝐽 ∈ Top ∧ 𝐴 ∈ 𝐽 ∧ 𝐵 ∈ 𝐽) → (𝐴 ∪ 𝐵) ∈ 𝐽)

Theorem

0opn 22879

The empty set is an open subset of any topology. (Contributed by Stefan Allan, 27-Feb-2006.)

⊢ (𝐽 ∈ Top → ∅ ∈ 𝐽)

Theorem

0ntop 22880

The empty set is not a topology. (Contributed by FL, 1-Jun-2008.)

⊢ ¬ ∅ ∈ Top

Theorem

topopn 22881

The underlying set of a topology is an open set. (Contributed by NM, 17-Jul-2006.)

⊢ 𝑋 = ∪ 𝐽    ⇒   ⊢ (𝐽 ∈ Top → 𝑋 ∈ 𝐽)

Theorem

eltopss 22882

A member of a topology is a subset of its underlying set. (Contributed by NM, 12-Sep-2006.)

⊢ 𝑋 = ∪ 𝐽    ⇒   ⊢ ((𝐽 ∈ Top ∧ 𝐴 ∈ 𝐽) → 𝐴 ⊆ 𝑋)

Theorem

riinopn 22883*

A finite indexed relative intersection of open sets is open. (Contributed by Mario Carneiro, 22-Aug-2015.)

⊢ 𝑋 = ∪ 𝐽    ⇒   ⊢ ((𝐽 ∈ Top ∧ 𝐴 ∈ Fin ∧ ∀𝑥 ∈ 𝐴 𝐵 ∈ 𝐽) → (𝑋 ∩ ∩ 𝑥 ∈ 𝐴 𝐵) ∈ 𝐽)

Theorem

rintopn 22884

A finite relative intersection of open sets is open. (Contributed by Mario Carneiro, 22-Aug-2015.)

⊢ 𝑋 = ∪ 𝐽    ⇒   ⊢ ((𝐽 ∈ Top ∧ 𝐴 ⊆ 𝐽 ∧ 𝐴 ∈ Fin) → (𝑋 ∩ ∩ 𝐴) ∈ 𝐽)

12.1.1.2  Topologies on sets

Syntax

ctopon 22885

Syntax for the function of topologies on sets.

class TopOn

Definition

df-topon 22886*

Define the function that associates with a set the set of topologies on it. (Contributed by Stefan O'Rear, 31-Jan-2015.)

⊢ TopOn = (𝑏 ∈ V ↦ {𝑗 ∈ Top ∣ 𝑏 = ∪ 𝑗})

Theorem

istopon 22887

Property of being a topology with a given base set. (Contributed by Stefan O'Rear, 31-Jan-2015.) (Revised by Mario Carneiro, 13-Aug-2015.)

⊢ (𝐽 ∈ (TopOn‘𝐵) ↔ (𝐽 ∈ Top ∧ 𝐵 = ∪ 𝐽))

Theorem

topontop 22888

A topology on a given base set is a topology. (Contributed by Mario Carneiro, 13-Aug-2015.)

⊢ (𝐽 ∈ (TopOn‘𝐵) → 𝐽 ∈ Top)

Theorem

toponuni 22889

The base set of a topology on a given base set. (Contributed by Mario Carneiro, 13-Aug-2015.)

⊢ (𝐽 ∈ (TopOn‘𝐵) → 𝐵 = ∪ 𝐽)

Theorem

topontopi 22890

A topology on a given base set is a topology. (Contributed by Mario Carneiro, 13-Aug-2015.)

⊢ 𝐽 ∈ (TopOn‘𝐵)    ⇒   ⊢ 𝐽 ∈ Top

Theorem

toponunii 22891

The base set of a topology on a given base set. (Contributed by Mario Carneiro, 13-Aug-2015.)

⊢ 𝐽 ∈ (TopOn‘𝐵)    ⇒   ⊢ 𝐵 = ∪ 𝐽

Theorem

toptopon 22892

Alternative definition of Top in terms of TopOn. (Contributed by Mario Carneiro, 13-Aug-2015.)

⊢ 𝑋 = ∪ 𝐽    ⇒   ⊢ (𝐽 ∈ Top ↔ 𝐽 ∈ (TopOn‘𝑋))

Theorem

toptopon2 22893

A topology is the same thing as a topology on the union of its open sets. (Contributed by BJ, 27-Apr-2021.)

⊢ (𝐽 ∈ Top ↔ 𝐽 ∈ (TopOn‘∪ 𝐽))

Theorem

topontopon 22894

A topology on a set is a topology on the union of its open sets. (Contributed by BJ, 27-Apr-2021.)

⊢ (𝐽 ∈ (TopOn‘𝑋) → 𝐽 ∈ (TopOn‘∪ 𝐽))

Theorem

funtopon 22895

The class TopOn is a function. (Contributed by BJ, 29-Apr-2021.)

⊢ Fun TopOn

Theorem

toponrestid 22896

Given a topology on a set, restricting it to that same set has no effect. (Contributed by Jim Kingdon, 6-Jul-2022.)

⊢ 𝐴 ∈ (TopOn‘𝐵)    ⇒   ⊢ 𝐴 = (𝐴 ↾_t 𝐵)

Theorem

toponsspwpw 22897

The set of topologies on a set is included in the double power set of that set. (Contributed by BJ, 29-Apr-2021.)

⊢ (TopOn‘𝐴) ⊆ 𝒫 𝒫 𝐴

Theorem

dmtopon 22898

The domain of TopOn is the universal class V. (Contributed by BJ, 29-Apr-2021.)

⊢ dom TopOn = V

Theorem

fntopon 22899

The class TopOn is a function with domain the universal class V. Analogue for topologies of fnmre 17544 for Moore collections. (Contributed by BJ, 29-Apr-2021.)

⊢ TopOn Fn V

Theorem

toprntopon 22900

A topology is the same thing as a topology on a set (variable-free version). (Contributed by BJ, 27-Apr-2021.)

⊢ Top = ∪ ran TopOn