Integration, IX, p. 115
Bourbaki marks this one as harder than the rest.
Notations and hypotheses are those of the preceding exercise. Let (μn) be a Cauchy sequence in M+b for the metric D; show that the sequence (μn) is convergent (this gives a new proof of the fact that the space M+b is Polish for the tight topology). One can proceed thus:
a) Let ε>0 and a>0 be two real numbers; there exists a finite subset F of T such that supnμn∙(T−Fa)≤ε (choose an integer N≥1 and a compact subset K of T such that supn≥ND(μn,μN)≤ε/2 and μN∙(T−K)≤ε/2; deduce from this that supn≥N∣μn∙(T)−μ∙(T)∣≤ε and supn≥Nμn∙(T−Ka/2)≤ε; finally, choose a finite subset F of T such that Ka/2⊂Fa and supn<Nμn∙(T−Fa)≤ε.
b) Let ε>0; there exists a compact subset K of T such that supμn∙(T−K)≤ε (for every integer p≥1 choose a finite subset Fp of T such that μn∙(T−(Fp)2−p)≤ε/2p, and set K=⋂p≥1(Fp)2−p).
c) Deduce from b) that the Cauchy sequence (μn) has a convergent subsequence (for the metric D).