scientific article; zbMATH DE number 1301699
Given \(N\) independent copies \(\xi_1(\cdot), \xi_2(\cdot), \ldots ,\xi_N(\cdot)\) of a diffusion process with time-dependent generator, the level-I empirical process \(\Sigma^N(t)\) is defined by \(\Sigma^N(t): = {1 \over N}\sum_{i=1}^N \delta_{\xi_i(t)}\). Given \(M\) independent copies \(\Sigma_1^N(t),\Sigma_2^N(t), \ldots , \Sigma_M^N(t) \) of the level-I empirical process \(\Sigma^N(\cdot)\), the level-II empirical process is defined by \(\Sigma^{M N}(t): = {1 \over M}\sum_{i=1}^M \delta_{\Sigma_i^N(t)}\). In [the authors, Probab. Theory Relat. Fields 98, No. 4, 423-487 (1994; Zbl 0794.60015)], a large deviation principle for the level-II empirical process was derived, for \(N, M \to \infty\). Here, the authors give an integral representation for the rate function which is analogous to the representation of the level-I rate function considered in [the authors, Stochastics 20, 247-308 (1987; Zbl 0613.60021)]. As an important tool, the authors introduce an appropriate class of distributions on spaces of measures and distribution-valued functionals and prove various approximation theorems for them. For the entire collection see [Zbl 0901.00049].
- Multilevel large deviations and interacting diffusions
- Large deviations for infinite dimensional analytical distributions
- Multidimensional large deviation local limit theorems
- scientific article; zbMATH DE number 4188872
- scientific article; zbMATH DE number 1374880
- Large exceedances for multidimensional Lévy processes
- Multivariate large deviations with stable limit laws
- Multilevel large deviations and interacting diffusions
- Multilevel mutation-selection systems and set-valued duals
- Large deviations and Nelson processes
- Large deviations for independent random variables – Application to Erdös-Renyi's functional law of large numbers
- Mean Field Analysis of Deep Neural Networks
- A random walk through Canadian contributions on empirical processes and their applications in probability and statistics
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 Q4246910)
