Skip to Content
Math4121Introduction to Lebesgue Integration (Lecture 26)

Math4121 Lecture 26

Lebesgue Measure

Lebesgue’s Integration

Partition on the y-axis, let ll be the minimum of f(x)f(x) on the yy-axis, LL be the maximum of f(x)f(x) on the yy-axis.

l=l0<l1<<ln=Ll=l_0<l_1<\cdots<l_n=L

Define Si={x[a,b]:li1f(x)<li}S_i=\{x\in[a,b]:l_{i-1}\leq f(x)<l_i\}

Defined the characteristic function of set SS is χS(x)=1\chi_S(x)=1 if xSx\in S and 00 otherwise.

Then ff lies between the following simple functions:

i=1nli1χSif(x)i=1nliχSi\sum_{i=1}^n l_{i-1}\chi_{S_i}\leq f(x)\leq \sum_{i=1}^n l_i\chi_{S_i}

This representation allows us to measure some weird sets on the xx-axis by constraining the yy-axis.

This is still kind of Riemann sum, but SiS_i can be very weird (not just intervals).

If we can “measure” each SiS_i, then we could define the integral of ff by

supl0,,lni=1nli1m(Si)infl0,,lni=1nlim(Si)\sup_{l_0,\cdots,l_n}\sum_{i=1}^n l_{i-1}m(S_i)\quad \inf_{l_0,\cdots,l_n}\sum_{i=1}^n l_i m(S_i)

If we used Jordan content, for mm here, this is just a different perspective of Riemann integral.

If we use Borel measure maybe things would be different (perhaps, better)?

As we discussed last time, this limits the measurable sets significantly.

Nonetheless, let’s try

  1. Characteristic function of the rational numbers
f(x)={1xQ[0,1]0otherwisef(x)=\begin{cases} 1 & x\in\mathbb{Q}\cap[0,1]\\ 0 & \text{otherwise} \end{cases}

Take partition 0=x0<x1<<xn=10=x_0<x_1<\cdots<x_n=1

S1={x:0f(x)<ϵ}=[0,1]QS_1=\{x:0\leq f(x)<\epsilon\}=[0,1]\setminus\mathbb{Q}

S2={x:ϵf(x)<1}=S_2=\{x:\epsilon\leq f(x)<1\}=\emptyset

S3={x:1f(x)<1+ϵ}=Q[0,1]S_3=\{x:1\leq f(x)<1+\epsilon\}=\mathbb{Q}\cap[0,1]

i=1nli1m(Si)=01+ϵ0+10=0\sum_{i=1}^n l_{i-1}m(S_i)=0\cdot 1+\epsilon\cdot 0+1\cdot 0=0 i=1nlim(Si)=ϵ1+10+(1+ϵ)0=ϵ\sum_{i=1}^n l_i m(S_i)=\epsilon\cdot 1+1\cdot 0+(1+\epsilon)\cdot 0=\epsilon

So, 001f(x)dmϵ0\leq\int_0^1 f(x)dm\leq\epsilon, here mm means the measure we used.

As ϵ\epsilon is arbitrary, we have 01f(x)dm=0\int_0^1 f(x)dm=0.

This shows that χS(x)dm=m(S)\int \chi_S(x)dm=m(S) for any measurable set SS.

Lebesgue’s Measure

Definition of Lebesgue measure

Outer Measure:

Given S[a,b]S\subset [a,b], let C\mathcal{C} be the collection of all countable covers of SS by open intervals.

me(S)=infCCm(C)m_e(S)=\inf_{C\in\mathcal{C}}m(C)

where m(C)m(C) is the Borel measure.

Recall such CC are Boreal measurable because open interval are in B\mathcal{B} and B\mathcal{B} (being a sigma algebra) is closed under countable unions.

Properties:

  1. Translation invariant: me(S+a)=me(S)m_e(S+a)=m_e(S)
  2. Countable additivity: me(ST)=me(S)+me(T)m_e(S\cup T)=m_e(S)+m_e(T) if ST=S\cap T=\emptyset
  3. m([0,1])=1m([0,1])=1

Notice we don’t have the difference property here.

Last updated on