Abstract:We revisit the question of whether the strong law of large numbers (SLLN) holds uniformly in a rich family of distributions, culminating in a distribution-uniform generalization of the Marcinkiewicz-Zygmund SLLN. These results can be viewed as extensions of Chung's distribution-uniform SLLN to random variables with uniformly integrable $q^\text{th}$ absolute central moments for $0 < q < 2$. Furthermore, we show that uniform integrability of the $q^\text{th}$ moment is both sufficient and necessary for the SLLN to hold uniformly at the Marcinkiewicz-Zygmund rate of $n^{1/q - 1}$. These proofs centrally rely on novel distribution-uniform analogues of some familiar almost sure convergence results including the Khintchine-Kolmogorov convergence theorem, Kolmogorov's three-series theorem, a stochastic generalization of Kronecker's lemma, and the Borel-Cantelli lemmas. We also consider the non-identically distributed case.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is about the uniformity of the Strong Law of Large Numbers (SLLN) in rich families of distributions. Specifically, the goals of the paper are:
1. **Improve the convergence rate**: When random variables have higher - order moments, can the convergence rate of the law of large numbers be improved? That is, for all \(\varepsilon>0\), can it be proved that:
\[
\lim_{m \to \infty} \sup_{P \in \mathcal{P}} \left( \sup_{k \ge m} \left| \frac{1}{k^{1/q}} \sum_{i = 1}^k (X_i - E_P(X)) \right| \ge \varepsilon \right)=0
\]
where \(1 < q < 2\), and under the \(P\)-uniform integrability condition of the \(q\)-th moment.
2. **Handle finite absolute moments**: When random variables have finite absolute moments that are less than 1 but more than 0, can their divergence rate be restricted? That is, for all \(\varepsilon>0\), can it be proved that:
\[
\lim_{m \to \infty} \sup_{P \in \mathcal{P}} \left( \sup_{k \ge m} \left| \frac{1}{k^{1/q}} \sum_{i = 1}^k X_i \right| \ge \varepsilon \right)=0
\]
where \(0 < q < 1\), and under the \(P\)-uniform integrability condition of the \(q\)-th moment, even if for some \(P\in\mathcal{P}\), \(E_P(|X|)=\infty\).
3. **Necessity condition**: Is the \(P\)-uniform integrability condition a necessary condition for the \(P\)-uniform law of large numbers to hold? That is, if the condition:
\[
\lim_{m \to \infty} \sup_{P \in \mathcal{P}} E_P\left( |X - E_P(X)|^q\cdot1_{\{|X - E_P(X)|^q > m\}} \right)=0
\]
does not hold, can it be proved that there exists some positive constant \(C > 0\) such that:
\[
\lim_{m \to \infty} \sup_{P \in \mathcal{P}} \left( \sup_{k \ge m} \left| \frac{1}{k} \sum_{i = 1}^k (X_i - E_P(X)) \right| \ge C \right)>0
\]
4. **Non - identically distributed case**: Is there a version similar to Chung's law of large numbers that is applicable to independent but non - identically distributed random variables? That is, can it be proved that:
\[
\lim_{m \to \infty} \sup_{P \in \mathcal{P}} \left( \sup_{k \ge m} \left| \frac{1}{a_k} \sum_{i = 1}^k (X_i - E_P(X_i)) \right| \ge \varepsilon \right)=0
\]
where \(a_n\to\infty\) is a properly chosen sequence, and holds for \((X_n)_{n = 1}^\infty\) under some \(P\)-uniform conditions.
The paper introduces new \(P\)-uniform almost sure convergence results (such as the \(P\)-uniform Khintchine - Kolmogorov convergence theorem, \(P\)-uniform Kolmogoro)