Bourbaki

Proposition 12

Tag 064O, permanent. Cite it as [Bourbaki, Tag 064O].

Algebra, chapter III, § 8. Determinants

For every invertible square matrix XX of type (n,n)(n, n), the inverse of XX is given by the formula

(27)(27)
X−1=(det⁡X)−1tY X^{-1} = (\det X)^{-1} {}^t Y

where YY is the cofactor matrix of XX.

By considering the transpose of XX and using Proposition 8 of no. 5, the Laplace expansions could be obtained relative to two complementary sets of rows and, in particular, the expansion of det⁡X\det X by a row, thus there are formulae equivalent to

(28)(28)
X.tY=(det⁡X)In, X . {}^t Y = (\det X) I_n,

in the above notation.

It is easily verified that if XX is the matrix of an endomorphism uu of a free A\mathbf{A}-module MM of dimension nn with respect to a basis (ei)1≤i≤n(e_i)_{1 \leq i \leq n}, tY{}^t Y is the matrix of the endomorphism u~\tilde{u} of MM defined by the following condition: for every set of nn elements x,y2,…,ynx, y_2, \ldots, y_n of MM,

u~(x)∧y2∧⋯∧yn=x∧u(y2)∧⋯∧u(yn). \tilde{u}(x) \wedge y_2 \wedge \cdots \wedge y_n = x \wedge u(y_2) \wedge \cdots \wedge u(y_n).

u~\tilde{u} is called the cotranspose of uu (cf. § 11, no. 11, Corollary to Proposition 13).

Bourbaki puts the proof directly after the statement, so what is above is the statement and its proof together.

Cited by

Nothing.

Cites

Transcribed from the printed edition, 1998, Springer. Not translated.