Fabio Rinaldi

Fabio Rinaldi is a Research Associate in the Optimal Policy and Applications Division at the Institute for Advanced Dynamic Uncertainty. He holds a PhD in Economics from Università Bocconi (Department of Economics), where his doctoral research developed Bayesian estimation methods for state-space representations of dynamic stochastic general equilibrium models. His thesis addressed the problem of posterior concentration in high-dimensional DSGE systems, establishing conditions under which Markov Chain Monte Carlo samplers converge reliably to the posterior distribution when the state-space dimension is large relative to the observable time series. The analysis produced efficient sequential Monte Carlo algorithms for filtering and smoothing in nonlinear state-space models, with applications to the estimation of latent productivity and monetary policy shock processes from central bank data. At IADU, Rinaldi develops the Bayesian estimation and filtering infrastructure that supports the empirical calibration of the Institute's stochastic control and mean field models. His work encompasses particle filtering for continuous-time diffusion models observed at discrete intervals, Bayesian model comparison for competing structural specifications, and time-series identification of the structural parameters that govern sovereign fiscal and monetary dynamics.