Bayesian inference for generalised tempered stable Levy processes.

• Applicant: Professor Dr. Denis Belomestny
• Subject Area: Mathematics
• Term: from 2018 to 2022
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 406700014

The goal of the project is the development of new efficient methods of Bayesian inference for Levy processes based on their discrete-time observations and theoretical investigation of these methods. In particular, for the class of generalized tempered stable processes, we plan to estimate the tempering function using a nonparametric Bayesian approach. An important task is development of efficient MCMC algorithms and proof of the corresponding contraction rates. The implementation of proposed methods and their application to financial and insurance data is also foreseen.

• DFG Programme: Research Grants
• International Connection: Netherlands
• Cooperation Partners: Dr. Shota Gugushvili; Professor Dr. Peter Spreij