The divergence Borel-Cantelli Lemma revisited (Q2097557): Difference between revisions

Let \((\Omega,A,\mu)\) be a probability space. The classical Borel-Cantelli Lemma states that for any sequence of \(\mu\)-measurable sets \(E_{i}\) \((i=1,2,3,\dots)\), if the sum of their measures converges then the corresponding lim sup set \(E_\infty\) is of measure zero. In general the converse statement is false. However, it is well known that the divergence counterpart is true under various additional independence hypotheses. In this work authors revisit these hypotheses and establish both sufficient and necessary conditions for \(E_\infty\) to have either positive or full measure based on the following lemmas: Convergence Borel-Cantelli Lemma. Let \((\Omega,A,\mu)\) be a probability space and let \(\{E_{i}\}_{i\in N}\) be a sequence of events in \(A\). Suppose that \(\Sigma\mu(E_{i})<+\infty\). Then \(\mu(E_\infty)=0\). Divergence Borel-Cantelli Lemma [\textit{G. Harman}, Metric number theory. Oxford: Clarendon Press (1998; Zbl 1081.11057); \textit{R. P. Bambah} (ed.) et al., Number theory. Basel: Birkhäuser (2000; Zbl 0935.00033); \textit{S. C. Port}, Theoretical probability for applications. Chichester: Wiley \& Sons (1994; Zbl 0860.60001); \textit{R. E. A. C. Paley} and \textit{A. Zygmund}, Proc. Camb. Philos. Soc. 26, 337--357, 458--474 (1930; JFM 56.0254.01); \textit{V. G. Sprindzhuk}, Metric theory of Diophantine approximations (Russian). Moskva: Izdatel'stvo ``Nauka'' (1977; Zbl 0417.10044); \textit{R. E. A. C. Paley} and \textit{A. Zygmund}, Proc. Camb. Philos. Soc. 28, 190--205 (1932; Zbl 0006.19802)]. Let \((\Omega,A,\mu)\) be a probability space and let \(\{E_{i}\}_{i\in N}\) be a sequence of events in \(A\). Suppose that \(\Sigma\mu(E_{i})=\infty\) and that there exists a constant \(C>0\) such that \[ \Sigma\mu(E_{s}\cap E_{t})\leq C(\Sigma\mu(E_{s}))^2 \] holds for infinitely many \(Q\in N\). Then \(\mu(E_\infty)\geq(1/C)\). In particular, if \(C=1\) then \(\mu(E_\infty)=1\). The following is a local variant of the standard divergence Borel-Cantelli Lemma which allows authors to deduce full measure rather than just positive measure. Local Borel-Cantelli Lemma. Let \((\Omega,A,\mu,d)\) be a metric measure space equipped with a Borel probability measure \(\mu\) and let \(\{E_{i}\}_{i\in N}\) be a sequence of Borel subsets of \(\Omega\). Suppose there exists an increasing function \(f:]0,+\infty[\to]0,+\infty[\) with \(f(x)\to 0\) as \(x\to 0\) such that for any open set \(A\) with \(\mu(A)>0\) there is a sequence \(\{L_{i,A}\}_{i\in N}\) of measurable subsets of \(A\) such that \(\Sigma\mu(L_{i,A})=\infty\) and for infinitely many \(Q\in N\) \[ \Sigma\mu(L_{s,A}\cap L_{t,A})\leq (1/(f(\mu(A))))(\Sigma\mu(L_{s,A}))^2 \] Then \(\mu(E_\infty)=1\). Moreover, if in addition \(\mu\) is doubling and \(f(x)=cx\) for some constant \(0<c\leq 1\), it suffices to take \(A\) in the above to be an arbitrary ball of sufficiently small radius centered in \(\operatorname{supp}\mu\). Some examples of application are given at the end of this work.