Key research themes
1. How can corrections beyond the quantum regression theorem improve the modeling of two-time correlation functions in non-Markovian open quantum systems?
This theme investigates advanced formulations for two-time correlation functions (TTCFs) in open quantum systems beyond the traditional quantum regression theorem (QRT). Whereas QRT provides a convenient link between one-time expectation values and two-time correlations assuming Markovianity, it fails to accurately describe TTCFs in non-Markovian (memory-retaining) environments typical of strong system-environment couplings or structured baths. Improved methods seek to incorporate memory effects, environmental noise influences, and temperature-dependent dynamics, thereby refining spectral predictions and coherence descriptions in quantum devices.
2. What is the structure and behavior of correlation functions and dependence measures in stochastic processes with long-range memory and non-Markovian properties?
This theme addresses the characterization and quantification of dependencies and correlation structures in stochastic processes exhibiting long memory and non-Markovian dynamics—phenomena common in physics, biology, economics, and network science. It unites spectral methods and generalized functional and operator frameworks (including spectral projections, copulas, and subordinated processes) to rigorously describe correlations beyond classical Markov assumptions, focusing on both covariance asymptotics and structural decompositions that enable capturing short-range to long-range dependent regimes and their statistical signatures.
3. How can advanced copula models based on convex combinations of Archimedean copulas improve the modeling of tail dependencies and asymmetric correlations in multivariate risk and dependence analysis?
This research direction develops composite copula models that integrate distinct Archimedean copulas such as Clayton, Gumbel, and Frank into convex combinations or mixtures to flexibly capture complex dependence structures including asymmetric tail behaviors in multivariate distributions. These hybrid models aim to overcome limitations of single copulas in modeling either lower or upper tail dependence exclusively, facilitating comprehensive risk assessment, dependence quantification, and inference in finance, insurance, hydrology, and other applied fields where asymmetric extremal dependence is crucial.