You are browsing the curriculum of a past academic year (2020–2021).
Do you want to change to the ongoing academic year?
Course unit, curriculum year 2020–2021
DATA.STAT.620
Bayesian filtering and smoothing, 5 cr
Tampere University
- Description
- Completion options
Teaching periods
Active in period 3 (1.1.2021–7.3.2021)
Active in period 4 (8.3.2021–31.5.2021)
Course code
DATA.STAT.620Language of instruction
EnglishAcademic year
2020–2021Level of study
Advanced studiesGrading scale
General scale, 0-5Persons responsible
Responsible teacher:
Robert PicheResponsible organisation
Faculty of Information Technology and Communication Sciences 100 %
Core content
- Multivariate probability basics and the multivariate Gaussian distribution.
- Kalman filter
- EKF, UKF, bootstrap particle filter
- Bayesian fixed-interval filtering, RTS smoother
- State-space model parameter estimation using MCMC
Complementary knowledge
- Chebyshev inequality
- Stationary Kalman filter, information filter, treatment of missing measurement
- EKF2, GHKF, importance sampling, SIR
- RTS extensions; particle smoother
- State space model parameter estimation using EM
Specialist knowledge
- Laws of total expectation and total variance
- discretisations of stochastic differential equation; Joseph formula
- stratified resampling, RB particle filter
- fixed-lag smoothing; fixed-point smoothing
Learning outcomes
Prerequisites
Compulsory prerequisites
Recommended prerequisites
Further information
Learning material
Equivalences
Studies that include this course
Completion option 1
Students self-study the textbook and recorded lectures, participate in weekly tutorials (discuss lecture material and present homework problem solutions), and do a 24-hour open-book take-home exam.
Participation in teaching
12.01.2021 – 30.04.2021
Active in period 3 (1.1.2021–7.3.2021)
Active in period 4 (8.3.2021–31.5.2021)