This is a continuation of the study of the book The Elements of Integration and Lebesgue Measure by Bartle, listing a few examples of functions that are measurable. Bartle is a very good author, he tries his very best to make this difficult subject accessible to undergraduates.
Example:
If X is the set R of real numbers, and X is the Borel algebra B, then any monotone function is Borel measurable.
Proof:
Suppose that f is monotone increasing, i.e. implies
.
Then, consists of a half-line which is either of the form
or the form
. (We will show later that both cases can occur.) Thus, the set will belong to the Borel algebra B which is the
-algebra generated by all open intervals (a,b) in R.
Both cases can indeed occur. For example, if f(x)=x, then the set will be of the form . More interestingly, if the set is the step function
, then when
, the set will be
.
Lemma: An extended real-valued function f is measurable if and only if the sets ,
belong to X and the real-valued function
defined by
is measurable.
This lemma is often useful when dealing with extended real-valued functions.
Proof: If f is in M(X,X), it is proven earlier in the book by Bartle that A and B belong to X. Let and
, then we have that
which is in X since it is the complement of the union of A and
.
If , then
, which is a union of two sets in X and hence also in X.
Hence, is measurable.
Conversely, if and
is measurable, then
when
, and
when
, due to a similar reason as above. Therefore f is measurable!