I'm trying to follow the proof of this fact on this book on Gröbner basis. At one point of the proof they use this equality that I do not fully understad. I wonder if it is evident:

Given prime monomial ideals $P_i$ (which are generated by sets of variables) in $K[X_1,\ldots,X_n]$ and another variable $x$ (that may or not be a generator in some ideals $P_i$) then I would need to show that $\langle x \rangle + \cap P_i = \cap (\langle x \rangle + P_i)$ where intersections are finite.

The inclusion $\subseteq$ is clear. For the other one I'm having more problems. The intuition is that each $\cap P_i$ is computed by taking $lcm(X_i,X_j) = X_iX_j$ and therefore consists of monomials. Can you help me to prove the other inclusion?

  • $\begingroup$ This is quite easy. First of all assume that you have only two prime ideals. If a monomial $m$ belongs to $(x,P_1)\cap(x,P_2)$ then it is divisible by $x$ (and you are done) or by some generator of $P_1$ and $P_2$. $\endgroup$ – user26857 Feb 2 '18 at 17:05

In general, if $J$ is a monomial ideal and $(I_i)_i$ is a family of monomial ideals of $K[x_1,\ldots,x_n]$ then $$ \bigcap_{i} (J+I_i) = J + \bigcap_i I_i.$$ This is easy to prove. The inclusion $\supseteq$ is clear. Conversely, let $u\in \bigcap_{i} (J+I_i)$ be a monomial. If $u\in J$ then there is nothing to prove. If $u\notin J$, since $u\in J+I_i$, then $u\in I_i$. Hence $u\in \bigcap_i I_i$.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.