scientific article; zbMATH DE number 1301692
Consider partial sums \(S_n= X_1+\cdots+ X_n\) \((n=1,2,\dots)\), \(S_0= 0\), of an i.i.d. sequence \(X_1,X_2,\dots\) with \(EX_1= 0\), \(EX^2_1= 1\), and finite moment generating function. Our main result proves that \[ \lim_{n\to\infty} \max_{0\leq j<n,1\leq k\leq n-j}(S_{j+k}- S_j)/k\alpha(k/\log n)= 1\quad\text{a.s.}, \] where \(\alpha\) denotes the inverse Chernoff function of \(X_1\). As a corollary, the above a.s. characterization of the maximum increments of \(\{S_n:n\geq 1\}\) contains an affirmative answer of Shao (1995) to a conjecture of Révész (1990). It also demonstrates that a proper normalization may depend on the full underlying distribution. An analogous result is presented for renewal counting processes based on nonnegative i.i.d. interarrival times (say) \(Y_1,Y_2,\dots\)\ . For the entire collection see [Zbl 0901.00049].
- Maxima of increments of partial sums for certain subexponential distributions
- A probabilistic representation of constants in Kesten's renewal theorem
- Limit results on finite-dimensional Gaussian random fields
- Rumor processes on \(\mathbb N\) and discrete renewal processes
- Two-dimensional renewal theorems with weak moment conditions and critical Bellman-Harris branching processes
- On a Conjecture of Revesz
- scientific article; zbMATH DE number 1143934 (Why is no real title available?)
- scientific article; zbMATH DE number 2118869 (Why is no real title available?)
- Extremes of the standardized Gaussian noise
- Shao's theorem on the maximum of standardized random walk increments for multidimensional arrays
- On Chung's LIL and Csáki's law for renewal processes
- Limiting distribution for the maximal standardized increment of a random walk
This page was built for publication:
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4246903)
