Master's degree in Statistics and Operations Research

Share:

MEIO - Summer School - Sequential Monte Carlo and MCMC methods applied to Stochastic Volatility Models

Títol del cursSequential Monte Carlo and MCMC methods applied to Stochastic Volatility Models
Impartit per
Hedibert Freitas Lopes (Associate Professor of Econometrics and Statistics, Graduate School of Business, The University of Chicago, Chicago, Illinois, USA)
Llengua del cursAnglès
Dates i horaris del curs
22,23,26,29 de juny i 1 de juliol de 2009, de 9 a 13:30h.
Tipus d'activitat i càrrega lectivaCurs de 20 hores
Reconeixement acadèmic
2,5 ECTS com a assignatures optatives per als estudiants del MEIO UPC-UB, com a ALE per als de PRIMER I SEGON CICLE i com crèdits pel DOCTORAT.
Data de matrículaDel 18 de maig al 17 de juny de 2009
Temari del curs1. Bayesian inference, decision and selection: a brief review
2. Monte Carlo (MC) methods
2.1 MC integration
2.2 Sampling importance resampling (SIR)
3. Markov chain MC (MCMC) methods
3.1 Gibbs sampler
3.2 Metropolis-Hastings algorithms
4. Stochastic volatility (SV) models
4.1 SV as a dynamic models
4.2 Posterior inference in dynamic models via MCMC schemes
4.3 Revisiting Jaquier, Polson and Rossi (1994) and Kim, Shephard and Chib (1994)
5. Sequential Monte Carlo
5.1 Sequential importance sampling (SIS) - Gordon, Salmond and Smith (1993)
5.2 Auxiliary particle filters (APF) - Pitt and Shephard (1999)
5.3 APF + parameter learning - Liu and West (2001)
5.4 Particle learning (PL) - Carvalho, Johannes, Lopes and Polson (2008)
5.5 Comparing SIS, APF and PL in the stochastic volatility context