TheoremBase

Induction on k extracts, by the real Bolzano-Weierstrass theorem applied to one bounded coordinate at a time and composition of index maps, a subsequence whose first k coordinates converge; coordinatewise convergence then gives convergence in RnR^n, and for a closed bounded set the limit cannot lie in the open complement.

Proof

Each result cited is universally quantified over the data in its own statement. No choice principle is used: the induction below proves an existential statement for each kk, and no sequence of choices is formed.

Let nn and dEd_{E} be as in the statement; (Rn,dE)(\mathbb{R}^{n},d_{E}) is a metric space by Rn\mathbb{R}^n with the Euclidean Distance Is a Complete Metric Space with Coordinatewise Convergence §metric, so dEd_{E} satisfies Metric Space §symmetric. Components are as in Euclidean Space Rn\mathbb{R}^n: n-Tuples of Real Numbers with Their Componentwise Operations §membership.

Clause subsequence. Let (x(m))(x^{(m)}) be a sequence in Rn\mathbb{R}^{n} whose set of terms is bounded in (Rn,dE)(\mathbb{R}^{n},d_{E}): there is a real MM with dE(u,v)≤Md_{E}(u,v)\le M for all terms u,vu,v.

Bounded coordinates. Let k∈[n]k\in[n] and m∈Nm\in\mathbb{N}. By Euclidean Space Rn\mathbb{R}^n: n-Tuples of Real Numbers with Their Componentwise Operations §operations, (x(m)−x(1))k=xk(m)−xk(1)(x^{(m)}-x^{(1)})_{k}=x^{(m)}_{k}-x^{(1)}_{k}, so by Properties of the Euclidean Norm on Rn\mathbb{R}^n: Square, Distance, Vanishing, Coordinate Bounds, Homogeneity, Triangle Inequalities and Expansion of a Square §coordinate and Properties of the Euclidean Norm on Rn\mathbb{R}^n: Square, Distance, Vanishing, Coordinate Bounds, Homogeneity, Triangle Inequalities and Expansion of a Square §distance,

∣xk(m)−xk(1)∣≤∥x(m)−x(1)∥=dE(x(m),x(1))≤M,|x^{(m)}_{k}-x^{(1)}_{k}|\le\lVert x^{(m)}-x^{(1)}\rVert=d_{E}(x^{(m)},x^{(1)})\le M,

hence ∣xk(m)∣≤Mk|x^{(m)}_{k}|\le M_{k} with Mk=M+∣xk(1)∣M_{k}=M+|x^{(1)}_{k}|, a real number by Closure Facts in an Ordered Field: Negatives, Differences, Quotients, Absolute Values, Powers, Natural Numbers, Halves and Minima §absolute-value. So for every strictly increasing σ:N→N\sigma:\mathbb{N}\to\mathbb{N} the real sequence m↦xk(σ(m))m\mapsto x^{(\sigma(m))}_{k} (a map N→R\mathbb{N}\to\mathbb{R} by Sets and Maps: Ordinary Notation §maps, its condition being predicative by Predicative Substitution of Values of Functions and Binary Operations, and Predicative Formulas for Negatives, Differences, Absolute Values, Distances, Reciprocals and Quotients in an Ordered Field §value-substitution) is bounded, with the bound MkM_{k}.

Induction on the number of coordinates. For k∈N0k\in\mathbb{N}_{0} with k≤nk\le n let S(k)S(k) be the statement: there is a strictly increasing map σ:N→N\sigma:\mathbb{N}\to\mathbb{N} such that for every i∈[k]i\in[k] the real sequence m↦xi(σ(m))m\mapsto x^{(\sigma(m))}_{i} is convergent in the sense of Convergent Sequences of Real Numbers §converges. We prove S(k)S(k) for every k≤nk\le n by induction from 00, The Natural Numbers and the Natural Numbers with Zero: Arithmetic, Order, Induction and Recursion §induction, applied to the class {k∈N0:k≤n⇒S(k)}\{k\in\mathbb{N}_{0}:k\le n\Rightarrow S(k)\}; it is formed by a predicative formula with set parameters only (nn and the sequence (x(m))(x^{(m)})), as S(k)S(k) quantifies over sets only (maps N→N\mathbb{N}\to\mathbb{N}, natural numbers and elements of R\mathbb{R}).

S(0)S(0): [0][0] is empty by Intervals of Natural Numbers §segment, and the identity map of N\mathbb{N} is strictly increasing.

Let k<nk<n and assume S(k)S(k), witnessed by σ\sigma. Let cc be the real sequence m↦xk+1(σ(m))m\mapsto x^{(\sigma(m))}_{k+1} (note k+1∈[n]k+1\in[n] since k<nk<n); it is bounded, as shown above, so by Completeness of the Real Numbers for Sequences: Monotone Convergence, the Bolzano-Weierstrass Theorem and Cauchy Sequences §bolzano-weierstrass it has a convergent subsequence, that is, there is a strictly increasing τ:N→N\tau:\mathbb{N}\to\mathbb{N} with c∘τc\circ\tau convergent. Let σ′=σ∘τ\sigma'=\sigma\circ\tau, strictly increasing by Monotone Sequences and Subsequences: Comparison of All Terms, Growth of the Indices, and Subsequences of Subsequences §composition. For i=k+1i=k+1, the sequence l↦xk+1(σ′(l))l\mapsto x^{(\sigma'(l))}_{k+1} is c∘τc\circ\tau, convergent. For i∈[k]i\in[k], the sequence e:m↦xi(σ(m))e:m\mapsto x^{(\sigma(m))}_{i} converges to some real LL by S(k)S(k); by Metric Spaces: Balls, the Metric Topology, and the Real and Complex Numbers as Metric Spaces §real-line it converges to LL in (R,dR)(\mathbb{R},d_{\mathbb{R}}); its subsequence e∘τe\circ\tau, which is l↦xi(σ′(l))l\mapsto x^{(\sigma'(l))}_{i}, converges to LL in (R,dR)(\mathbb{R},d_{\mathbb{R}}) by the background of Metric Spaces: Balls, the Metric Topology, and the Real and Complex Numbers as Metric Spaces §background, hence converges to LL by Metric Spaces: Balls, the Metric Topology, and the Real and Complex Numbers as Metric Spaces §real-line. As i∈[k+1]i\in[k+1] means i∈[k]i\in[k] or i=k+1i=k+1, σ′\sigma' witnesses S(k+1)S(k+1).

Conclusion. By S(n)S(n) there is a strictly increasing σ:N→N\sigma:\mathbb{N}\to\mathbb{N} such that, for every i∈[n]i\in[n], the real sequence m↦xi(σ(m))m\mapsto x^{(\sigma(m))}_{i} is convergent, hence convergent in (R,dR)(\mathbb{R},d_{\mathbb{R}}) by Metric Spaces: Balls, the Metric Topology, and the Real and Complex Numbers as Metric Spaces §real-line. For i∈[n]i\in[n] let yi=lim⁡m→∞xi(σ(m))y_{i}=\lim_{m\to\infty}x^{(\sigma(m))}_{i}, its limit in (R,dR)(\mathbb{R},d_{\mathbb{R}}), unique by the same background clause, to which the sequence converges by The Limit of a Convergent Sequence in a Metric Space §converges-to-limit. Let y∈Rny\in\mathbb{R}^{n} be the map i↦yii\mapsto y_{i} on [n][n], given by Sets and Maps: Ordinary Notation §maps and Euclidean Space Rn\mathbb{R}^n: n-Tuples of Real Numbers with Their Componentwise Operations §membership. The subsequence (x(σ(m)))m∈N(x^{(\sigma(m))})_{m\in\mathbb{N}} of (x(m))(x^{(m)}), per Subsequences §subsequence, has kk-th coordinate sequence m↦xk(σ(m))m\mapsto x^{(\sigma(m))}_{k} converging to yky_{k} in (R,dR)(\mathbb{R},d_{\mathbb{R}}) for every k∈[n]k\in[n], so it converges to yy in (Rn,dE)(\mathbb{R}^{n},d_{E}) by Rn\mathbb{R}^n with the Euclidean Distance Is a Complete Metric Space with Coordinatewise Convergence §coordinatewise, and it is convergent.

Clause sequentially-compact. Let K⊆RnK\subseteq\mathbb{R}^{n} be closed in (Rn,TdE)(\mathbb{R}^{n},\mathcal{T}_{d_{E}}) and bounded in (Rn,dE)(\mathbb{R}^{n},d_{E}), with a real MM such that dE(u,v)≤Md_{E}(u,v)\le M for all u,v∈Ku,v\in K. Let (xm)(x_{m}) be a sequence in Rn\mathbb{R}^{n} with xm∈Kx_{m}\in K for every m∈Nm\in\mathbb{N}. Its set of terms is a subset of KK, hence bounded with the same MM by Bounded Subset of a Metric Space §bounded. By the clause subsequence, proved above, there are a strictly increasing σ:N→N\sigma:\mathbb{N}\to\mathbb{N} and x∈Rnx\in\mathbb{R}^{n} such that (xσ(k))k∈N(x_{\sigma(k)})_{k\in\mathbb{N}} converges to xx in (Rn,dE)(\mathbb{R}^{n},d_{E}).

We show x∈Kx\in K. Suppose x∉Kx\notin K. Since KK is closed, Rn∖K∈TdE\mathbb{R}^{n}\setminus K\in\mathcal{T}_{d_{E}} by Closed Subset of a Topological Space §closed, so Rn∖K\mathbb{R}^{n}\setminus K is open in (Rn,dE)(\mathbb{R}^{n},d_{E}) by Metric Spaces: Balls, the Metric Topology, and the Real and Complex Numbers as Metric Spaces §topology. In this order: as x∈Rn∖Kx\in\mathbb{R}^{n}\setminus K, Open Subset of a Metric Space §open gives a real r>0r>0 with BdE(x,r)⊆Rn∖KB_{d_{E}}(x,r)\subseteq\mathbb{R}^{n}\setminus K; then Convergent Sequence in a Metric Space §converges with ε=r\varepsilon=r gives N∈NN\in\mathbb{N} with dE(xσ(k),x)<rd_{E}(x_{\sigma(k)},x)<r for every k≥Nk\ge N. In particular dE(x,xσ(N))=dE(xσ(N),x)<rd_{E}(x,x_{\sigma(N)})=d_{E}(x_{\sigma(N)},x)<r, so xσ(N)∈BdE(x,r)x_{\sigma(N)}\in B_{d_{E}}(x,r) by Open Ball in a Metric Space §membership, whence xσ(N)∉Kx_{\sigma(N)}\notin K, contradicting xσ(N)∈Kx_{\sigma(N)}\in K. Hence x∈Kx\in K, and KK is sequentially compact in (Rn,dE)(\mathbb{R}^{n},d_{E}).

Citations

Loading…

Dependencies

Uses0

Loading…

Comments

Log in to comment.

Loading…