Overview
Siem Jan Koopman is a time-series econometrician (Vrije Universiteit Amsterdam; Tinbergen Institute), a leading authority on state-space methods — Kalman filtering and smoothing, unobserved-components and structural time-series models, simulation smoothing, and stochastic volatility. He is co-author, with James Durbin, of the standard reference Time Series Analysis by State Space Methods.
Key Contributions / Features
- Exact diffuse initialization (Koopman 1997): the analytically exact solution for initializing the Kalman filter and smoother for nonstationary (diffuse-initial-condition) models.
- Time Series Analysis by State Space Methods (Durbin-Koopman 2001): the standard graduate/reference text.
- Simulation smoothing (Durbin-Koopman 2002): a simple, efficient simulation smoother for Bayesian and simulated-likelihood state-space analysis.
- State-space software and SV (Koopman-Shephard-Doornik, SsfPack; Sandmann-Koopman, Monte Carlo maximum-likelihood estimation of stochastic volatility models).
Related