Bourbaki

Exercise 9, § 4

Commutative Algebra, VIII, p. 92

On uses the notations of the preceding exercise. We put on M\mathscr{M} the lexicographic order. Let l∈Nl \in \mathbf{N}. A subset A of Ml\mathscr{M}_l is said to be initial if, for every m∈Am \in A, the set

[m]={m′∈Md(m)∣m′<m} [m] = \{ m' \in \mathscr{M}_{d(m)} | m' < m \}
is contained in A. (a) Let l∈Nl \in \mathbf{N} (resp. l∈N−{0}l \in \mathbf{N} - \{0\}) and A⊂MlA \subset \mathscr{M}_l be an initial subset. Now prove that δA\delta A (resp. dAdA) is an initial subset of Ml+1\mathscr{M}_{l+1} (resp. Ml−1\mathscr{M}_{l-1}). Suppose that A possesses a greatest element m=X1a1...Xnanm = X_1^{a_1} ... X_n^{a_n}. Determine the greatest element of δA\delta A (resp. dAdA). (b) Let l∈N−{0}l \in \mathbf{N} - \{0\} and A⊂MlA \subset \mathscr{M}_l be a finite nonempty initial subset. Let XalX_{a_l} be the largest variable such that [Xall]⊂A[X_{a_l}^l] \subset A. Now prove that every mm in A−[Xall]A - [X_{a_l}^l] is equal to 1 or divisible by Xal+1X_{a_l+1} and that A′={m∈Ml−1∣mXal+1∈A−[Xall]}A' = \{ m \in \mathscr{M}_{l-1} | m X_{a_l+1} \in A - [X_{a_l}^l] \} is an initial subset of Ml−1\mathscr{M}_{l-1}. Deduce from what precedes an expression of Card([X1b1...Xnbn])\mathrm{Card}([X_1^{b_1} ... X_n^{b_n}]). (c) Deduce from what precedes that, for every l∈N−{0}l \in \mathbf{N} - \{0\} and every finite initial subset A of Ml\mathscr{M}_l, one has Card(δA)=∂l(Card(A))\mathrm{Card}(\delta A) = \partial_l(\mathrm{Card}(A)), where ∂l\partial_l was defined in the preceding exercise. Calculate in an analogous manner Card(dA)\mathrm{Card}(dA). (d) For every l∈Nl \in \mathbf{N} and every n∈Nn \in \mathbf{N}, put
μ(n)=inf⁡A⊂MlCard(A)=nCard(dA), \mu(n) = \inf_{\substack{A \subset \mathscr{M}_l \\ \mathrm{Card}(A) = n}} \mathrm{Card}(dA),
and denote by mnm_n the n-th monomial of Ml\mathscr{M}_l for the lexicographic order (one will put m0=1m_0 = 1).

For every finite subset A of Ml\mathscr{M}_l, denote by CACA the initial subset of Ml\mathscr{M}_l such that Card(A)=Card(CA)\mathrm{Card}(A) = \mathrm{Card}(CA). Now prove that the following three assertions are equivalent: (MC1) for every l∈Nl \in \mathbf{N} and every A⊂MlA \subset \mathscr{M}_l, one has CδA⊂δCAC\delta A \subset \delta CA; (MC2) for every l∈Nl \in \mathbf{N} and every A⊂MlA \subset \mathscr{M}_l, one has dCA⊂CdAdCA \subset CdA; (MC3) for every l∈Nl \in \mathbf{N} and every n∈Nn \in \mathbf{N}, one has Card(d[mn])=μ(n)\mathrm{Card}(d[m_n]) = \mu(n).

Solution

No solution has been written yet.

Machine translation of the English, by gpt-5-6-mini. Not checked by a person. That model is a cut down version of the one the rest of this language was written by.