Print Email Facebook Twitter A non-parametric Bayesian approach to decompounding from high frequency data Title A non-parametric Bayesian approach to decompounding from high frequency data Author Gugushvili, Shota (Universiteit Leiden) van der Meulen, F.H. (TU Delft Statistics) Spreij, Peter (Universiteit van Amsterdam; Radboud Universiteit Nijmegen) Date 2016 Abstract Given a sample from a discretely observed compound Poisson process, we consider non-parametric estimation of the density f0 of its jump sizes, as well as of its intensity λ0. We take a Bayesian approach to the problem and specify the prior on f0 as the Dirichlet location mixture of normal densities. An independent prior for λ0 is assumed to be compactly supported and to possess a positive density with respect to the Lebesgue measure. We show that under suitable assumptions the posterior contracts around the pair (λ0,f0) at essentially (up to a logarithmic factor) the nΔ−−−√-rate, where n is the number of observations and Δ is the mesh size at which the process is sampled. The emphasis is on high frequency data, Δ→0, but the obtained results are also valid for fixed Δ. In either case we assume that nΔ→∞. Our main result implies existence of Bayesian point estimates converging (in the frequentist sense, in probability) to (λ0,f0) at the same rate. We also discuss a practical implementation of our approach. The computational problem is dealt with by inclusion of auxiliary variables and we develop a Markov chain Monte Carlo algorithm that samples from the joint distribution of the unknown parameters in the mixture density and the introduced auxiliary variables. Numerical examples illustrate the feasibility of this approach. Subject Compound Poisson processNon-parametric Bayesian estimationPosterior contraction rateHigh frequency observations To reference this document use: http://resolver.tudelft.nl/uuid:20839fb6-0459-42cf-bfdd-05c71e48dc69 DOI https://doi.org/10.1007/s11203-016-9153-1 ISSN 1387-0874 Source Statistical Inference for Stochastic Processes, 1-27 Part of collection Institutional Repository Document type journal article Rights © 2016 Shota Gugushvili, F.H. van der Meulen, Peter Spreij Files PDF 10.1007_s11203_016_9153_1.pdf 11.05 MB Close viewer /islandora/object/uuid:20839fb6-0459-42cf-bfdd-05c71e48dc69/datastream/OBJ/view