# quotient of finitely presented module

Assume the exact sequence $$$$A \xrightarrow{f} B \xrightarrow{} Cokerf \xrightarrow{} o$$$$ where $A$ is a finitely generated module and $B$ is a finitely presented module. Is it true that $Cokerf$ is finitely presented module ? In general, is the quotient of a finitely presented module to a finitely generated submodule, finitely presented ?

-
What precisely are the categorical definitions of finitely generated, exact sequence, finitely presented? You might also throw in the definition of projective and the knowledge that free modules are projective. – Barbara Osofsky Feb 20 '13 at 3:42
This is easy to see if $B$ is projective. I have not even been able to find the answer online for the general case though. – Alex Youcis Feb 20 '13 at 5:01

This is true.

I'll denote the ring by $R$, while remaining silent about whether these are left or right modules.

Write $B = R^n/I$ for some finitely generated submodule $I \subseteq R^n$. Then the image of $f \colon A \to B$ is of the form $J/I$ for some submodule $J \subseteq R^n$, and $J/I$ is a finitely generated module. Then $J$ is finitely generated as well: taking finitely many elements of $J$ whose images generate $J/I$, along with a finite generating set for $I$, will give a finite generating set for $J$.

Thus $Coker(f) \cong R^n/J$ is finitely presented.

If you're looking for a reference, see Exercise 4.8 in T.Y. Lam's book Exercises in Modules and Rings.

-
The same proof shows: If $A \to B \to C \to 0$ is exact, $A$ is of finite type and $B$ is of finite presentation, then $C$ is of finite presentation. – Martin Brandenburg Sep 17 '13 at 9:55
@MartinBrandenburg, indeed! – Manny Reyes Sep 17 '13 at 13:13