Let $C$ be any collection of subsets of a set $X$. Let $A(C)$ the algebra generated by $C$. Prove that the $\sigma$-algebra generated by $C$ and $A(C)$ are the same.
Here is my idea of a proof,
Proof: I would like to show that one is a subset of another and vice-versa.
Let $\sigma(C)$ be the $\sigma$-algebra generated by $C$ and $\sigma [A(C)]$ be the $\sigma$-algebra generated by $A(C)$.
Since $C\subset A(C)$ and $A(C)\subset\sigma [A(C)]$ then $C\subset \sigma [A(C)]$. $\sigma(C)$ is the smallest $\sigma$-algebra containing $C$ thus $\sigma(C)\subset\sigma [A(C)]$.
For the other way, note that $C\subset \sigma(C)$. Since $\sigma(C)$ is also an algebra thus $A(C)\subset \sigma(C)$. $\sigma [A(C)]$ is the smallest algebra containing $[A(C)]$ thus $\sigma [A(C)]\subset \sigma(C)$.
Is my proof correct? In the second part I assumed that $A(C)$ is the smallest algebra containing $C$ that is why I got $A(C)\subset \sigma(C)$. I can't find the definition of "the algebra generated by $C$" anywhere. The definition I have is that the $\sigma$-algebra generated by a set $C$ is the smallest $\sigma$-algebra that contains $C$. So I assumed that "the algebra generated by $C$" is the smallest algebra that contains $C$. Am I right in assuming this? I hope somebody could clarify this for me.
Also, what is the smallest algebra containing a set $C$? Is it $\{\emptyset,C\}$? So is $\{\emptyset,C\}$ the algebra generated by $C$?
Thank you!
Solved 1 Answer
See More Answers for FREE
Enhance your learning with StudyX
Receive support from our dedicated community users and experts
See up to 20 answers per week for free
Experience reliable customer service
That's right. Whenever you have "the $A$ generated by $B$", it almost always means "the smallest $A$ containing $B$" in some sense (usually there's some large object which his also $A$ and contains $B$ so that the statement even makes sense, or some adjective describing the way the object is constructed, that is, we know what operations on $B$ look like; in your case, the large object is the algebra $\mathcal P(X)$ of all subsets of $X$, but on the other hand, we don't need an universe to know what a ring of sets generated by a given family looks like, neither do we need that to say what a group freely generated by a set looks like).For your second point, it would depend on the universe of the algebra. If $C=X$ is the universe, then the algebra is $\{\emptyset,C\}$. In general, it is $\{\emptyset,C,X\setminus C, X\}$ where $X$ is the universe. ...