TheoremBase

The cubes have finite measure and cover ℝⁿ; a bounded set lies in one cube; a nonempty open set contains a nonempty open box; and the part of a coordinate hyperplane inside a cube lies in boxes of arbitrarily small volume.

Proof

Open boxes lie in B(Rn)\mathcal{B}(\mathbb{R}^{n}) (Existence and Uniqueness of Lebesgue Measure on ℝⁿ (i), Closure Properties of σ-Algebras, Intersections and Traces of σ-Algebras, and the Generated σ-Algebra (iii)), and the cubes QkQ_{k}, k∈Nk\in\mathbb{N}, cover Rn\mathbb{R}^{n} (Existence and Uniqueness of Lebesgue Measure on ℝⁿ (ii)). For a,b∈Rna,b\in\mathbb{R}^{n} with ai<bia_{i}<b_{i} for every i∈[n]i\in[n],

λn(R(a,b))=∏i=1n(bi−ai)\lambda_{n}\bigl(R(a,b)\bigr)=\prod_{i=1}^{n}(b_{i}-a_{i})

(Lebesgue Measure on Rn\mathbb{R}^n (i), Existence and Uniqueness of Lebesgue Measure on ℝⁿ (iii)); in particular λn(Qk)=∏i=1n2k\lambda_{n}(Q_{k})=\prod_{i=1}^{n}2k is a real number, so λn(Qk)<∞\lambda_{n}(Q_{k})<\infty.

σ\sigma-finiteness. (Qk)k∈N(Q_{k})_{k\in\mathbb{N}} is a sequence in B(Rn)\mathcal{B}(\mathbb{R}^{n}) with union Rn\mathbb{R}^{n} and λn(Qk)<∞\lambda_{n}(Q_{k})<\infty for every kk, so λn\lambda_{n} is σ\sigma-finite (Measure, Measure Space, and Probability Measure (iv)). This proves (i).

Bounded sets. Let B∈B(Rn)B\in\mathcal{B}(\mathbb{R}^{n}) be bounded, with dE(x,y)≤Md_{E}(x,y)\le M for all x,y∈Bx,y\in B (Bounded Subset of a Metric Space (i)). If B=∅B=\emptyset, then λn(B)=0\lambda_{n}(B)=0. Otherwise fix x0∈Bx^{0}\in B and choose k∈Nk\in\mathbb{N} with M+∥x0∥<kM+\lVert x^{0}\rVert<k (The Integers in the Real Numbers: Greatest Elements, the Integer Part, and Archimedean Consequences (ii)). For x∈Bx\in B and i∈[n]i\in[n],

∣xi∣≤∥x∥≤∥x−x0∥+∥x0∥=dE(x,x0)+∥x0∥≤M+∥x0∥<k|x_{i}|\le\lVert x\rVert\le\lVert x-x^{0}\rVert+\lVert x^{0}\rVert=d_{E}(x,x^{0})+\lVert x^{0}\rVert\le M+\lVert x^{0}\rVert<k

(Properties of the Euclidean Norm on Rn\mathbb{R}^n: Square, Distance, Vanishing, Coordinate Bounds, Homogeneity, Triangle Inequalities and Expansion of a Square (iv), Properties of the Euclidean Norm on Rn\mathbb{R}^n: Square, Distance, Vanishing, Coordinate Bounds, Homogeneity, Triangle Inequalities and Expansion of a Square (vi), Properties of the Euclidean Norm on Rn\mathbb{R}^n: Square, Distance, Vanishing, Coordinate Bounds, Homogeneity, Triangle Inequalities and Expansion of a Square (ii)). So B⊆QkB\subseteq Q_{k}, and λn(B)≤λn(Qk)<∞\lambda_{n}(B)\le\lambda_{n}(Q_{k})<\infty (Basic Properties of a Measure (iii)). This proves (ii).

Open sets. Let U⊆RnU\subseteq\mathbb{R}^{n} be open and nonempty. Then U∈B(Rn)U\in\mathcal{B}(\mathbb{R}^{n}) (Borel Sigma-Algebra of a Metric Space (i), Closure Properties of σ-Algebras, Intersections and Traces of σ-Algebras, and the Generated σ-Algebra (iii)). Fix x∈Ux\in U. By The Borel Sets of ℝⁿ and of the Real Matrices Are Generated by the Coordinates (i), UU is the union of a sequence of rational open boxes contained in UU, so x∈W⊆Ux\in W\subseteq U for one of them, W=R(a,b)W=R(a,b) with a,ba,b having rational, hence real, components. As ai<xi<bia_{i}<x_{i}<b_{i}, we have bi−ai>0b_{i}-a_{i}>0 for every i∈[n]i\in[n], so the product ∏i=1n(bi−ai)\prod_{i=1}^{n}(b_{i}-a_{i}) is nonnegative and nonzero (Products over Finite Sets: a Single Factor, Vanishing in a Field, and Signs and Monotonicity in an Ordered Ring (iii), Products over Finite Sets: a Single Factor, Vanishing in a Field, and Signs and Monotonicity in an Ordered Ring (ii)). Hence

λn(U)≥λn(W)=∏i=1n(bi−ai)>0\lambda_{n}(U)\ge\lambda_{n}(W)=\prod_{i=1}^{n}(b_{i}-a_{i})>0

(Basic Properties of a Measure (iii)). This proves (iii).

Coordinate hyperplanes. The set HH is the preimage of {c}=[c,c]∈B(R)\{c\}=[c,c]\in\mathcal{B}(\mathbb{R}) under the ii-th coordinate function, so H∈B(Rn)H\in\mathcal{B}(\mathbb{R}^{n}) (Intervals Are Borel, and Rays, Half-Open Intervals and Rational Intervals Generate the Borel Sigma-Algebra of the Real Line (ii), The Borel Sets of ℝⁿ and of the Real Matrices Are Generated by the Coordinates (iii), Closure Properties of σ-Algebras, Intersections and Traces of σ-Algebras, and the Generated σ-Algebra (iii)), and so is each H∩QkH\cap Q_{k} (Closure Properties of σ-Algebras, Intersections and Traces of σ-Algebras, and the Generated σ-Algebra (ix)). Fix k∈Nk\in\mathbb{N} and put C=∏l=1n2kC=\prod_{l=1}^{n}2k, a positive real number (Products over Finite Sets: a Single Factor, Vanishing in a Field, and Signs and Monotonicity in an Ordered Ring (iii), Products over Finite Sets: a Single Factor, Vanishing in a Field, and Signs and Monotonicity in an Ordered Ring (ii)). For j∈Nj\in\mathbb{N} let a,b∈Rna,b\in\mathbb{R}^{n} have components al=−ka_{l}=-k, bl=kb_{l}=k for l≠il\neq i, and ai=c−k/ja_{i}=c-k/j, bi=c+k/jb_{i}=c+k/j. Every x∈H∩Qkx\in H\cap Q_{k} has xi=cx_{i}=c and −k<xl<k-k<x_{l}<k for all ll, so H∩Qk⊆R(a,b)H\cap Q_{k}\subseteq R(a,b). Moreover bl−al=2k wlb_{l}-a_{l}=2k\,w_{l}, where wl=1w_{l}=1 for l≠il\neq i and wi=1/jw_{i}=1/j, so

λn(H∩Qk)≤λn(R(a,b))=∏l=1n2k wl=(∏l=1n2k)(∏l=1nwl)=Cj\lambda_{n}(H\cap Q_{k})\le\lambda_{n}\bigl(R(a,b)\bigr)=\prod_{l=1}^{n}2k\,w_{l}=\Bigl(\prod_{l=1}^{n}2k\Bigr)\Bigl(\prod_{l=1}^{n}w_{l}\Bigr)=\frac{C}{j}

(Basic Properties of a Measure (iii), Iterated Operations: Recursion, Splitting, Reordering, Termwise Combination and Homomorphisms (v); ∏l=1nwl=wi=1/j\prod_{l=1}^{n}w_{l}=w_{i}=1/j by Products over Finite Sets: a Single Factor, Vanishing in a Field, and Signs and Monotonicity in an Ordered Ring (i), all other factors being 11). In particular u=λn(H∩Qk)≤C/ju=\lambda_{n}(H\cap Q_{k})\le C/j is a real number. If uu were positive, there would be j∈Nj\in\mathbb{N} with 1/j<u/C1/j<u/C (The Integers in the Real Numbers: Greatest Elements, the Integer Part, and Archimedean Consequences (iii)), contradicting u≤C/ju\le C/j. So λn(H∩Qk)=0\lambda_{n}(H\cap Q_{k})=0 for every kk, and since H=⋃k∈N(H∩Qk)H=\bigcup_{k\in\mathbb{N}}(H\cap Q_{k}),

λn(H)=0\lambda_{n}(H)=0

(Null Sets and Properties Holding Almost Everywhere (iv)). This proves (iv).

Citations

Loading…

Dependencies

Uses0

Loading…

Comments

Log in to comment.

Loading…