On this site (under construction), I summarise some popular algorithms found in literature on state space models, Bayesian time series econometrics and econometrics in general. I also provide some implementation-ready code based on the object-oriented matrix language OxMetrics of Doornik (2007) and the library of state space functions SsfPack of Koopman et al. (2008).
- Linear and Gaussian state space model and the Kalman filter, the Kalman smoother and two simulation smoothers
- The basic stochastic volatility model (via 7-component mixture and the multi-move sampler)
- Time-varying parameter regression model with stochastic volatility
- Stochastic volatility in mean model (with time-varying coefficients)
- The bootstrap, auxiliary and tempered particle filter
- Particle Markov chain Monte Carlo and integrated likelihood via particle filter
- Particle Gibbs (with efficient importance sampling, with ancestor sampling)
- Interweaving strategy in Meteropolis-Hastings within Gibbs sampler
- The basic Hamiltonian Markov chain Monte Carlo
Reference:
Doornik, J.A. 2007. Object-Oriented Matrix Programming Using Ox, third ed. Timberlake Consultants Press, London
Koopman, S. J., Shephard, N., Doornik, J. A. 2008. Statistical Algorithms for Models in State Space Form: SsfPack 3.0. Timberlake Consultants Press, London
