Con preghiera di diffusione a potenziali interessati.
Grazie Matteo Ruggiero
--- Matteo Ruggiero University of Torino and Collegio Carlo Alberto www.matteoruggiero.it http://www.matteoruggiero.it/
Postdoctoral position at the University of Torino
Title: FAST INFERENTIAL STRATEGIES WITH COX-INGERSOLL-ROSS DRIVEN HIDDEN MARKOV MODELS AND THEIR EXTENSIONS Supervisor: Matteo Ruggiero (University of Torino and Collegio Carlo Alberto) www.matteoruggiero.it http://www.matteoruggiero.it/
Brief description: The project aims at improving current inferential techniques addressed to diffusions like Cox-Ingersoll-Ross processes or their extensions, e.g. Dawson-Watanabe processes, or transformations, e.g., Wright-Fisher and Fleming-Viot diffusions. We will tackle Bayesian inference on the trajectory of such diffusions or on the parameters that characterize drift and volatility, under the assumption of a hidden Markov model framework with noisy data collection. We will typically exploit duality properties, which allow to write the conditional distributions of the hidden signal given the data in simpler forms than those obtained through the transition density of the signal, especially when the latter is only available as a series expansion, together with simulation and approximation strategies, which aim at accelerating the inference by leveraging on the discrete nature of the dual process state space. Such approaches will be used to make direct inference on the process trajectory and to set up Markov chain Monte Carlo strategies that target the estimation of the process parameters.
Duration: 12 months, renewable. Application deadline: h13 Oct 24, 2022 Starting date: Dec 1 or later, 2022
Further info: matteo.ruggiero@unito.it Applications through https://www.serviziweb.unito.it/albo_ateneo/ https://www.serviziweb.unito.it/albo_ateneo/ Reference number for the call: 4382; code reference in the call: ESOMAS.2022.08/XXIV.