ST816: Computational Statistics (10 ECTS)

STADS: 25003601

Level
Master's level course

Teaching period
The course is offered in the spring semester.

Teacher responsible
Email: yuri.goegebeur@stat.sdu.dk

Timetable
Show entire timetable
Show personal time table for this course.

Comment:
Ubegrænset deltagerantal. Fælles undervisning med ST516,ST505,ST522,ST804,ST810.

Prerequisites:
None

Academic preconditions:
The content of the course ST521 Mathematical Statistics must be known.

Course introduction
To introduce modern computer intensive statistical methods as tools to investigate stochastic phenomena and statistical procedures, and to perform statistical inference. The participants will learn to use the statistical package R to carry out statistical analyses.

Expected learning outcome
After having followed the course the student should be able to

• reproduce key theoretical results concerning elementary operations on random variables and vectors, and to apply these to simple theoretical assignments, 
• reproduce and apply the fundamental theorems of random variate generation, 
• simulate variates and vectors from the most common distributions,
• evaluate the quality of a random number generator, 
• apply the basic principles of variance reduction, 
• simulate complex systems and investigate their properties, 
• use simulation to approximate integrals, 
• use simulation to compute p-values and confidence intervals, 
• investigate properties of statistical procedures and estimators using simulation, 
• perform programming relevant to the content of the course in the statistical package used in the course, 
• identify and interpret relevant information in the output of the statistical package used in the course, 
• summarize the results of an analysis in a statistical report.

Subject overview
Random number generators, inversion method, rejection sampling, simulation from multivariate distributions, Markov Chain Monte Carlo methods, permutation and randomization tests, transformations, simulation of experiments and complex systems, Monte Carlo integration, simulation of stochastic processes, bootstrap methods, Bayesian models and methods, EM algorithm, nonparametric density estimation.

Literature

Meddeles ved kursets start.

Website
This course uses e-learn (blackboard).

Prerequisites for participating in the exam
None

Assessment and marking:

1. The first part of the course is evaluated by projects. Danish 7 point scale, internal censor (5 ECTS)
2. The second part of the course is evaluated at an oral exam. Danish 7 point scale, internal censor. The oral exam is based on, but not limited to, projects made during the second half of the course (5 ECTS)

Expected working hours
The teaching method is based on three phase model.
Intro phase: 56 hours
Skills training phase: 36 hours, hereof:
 - Tutorials: 10 hours
 - Laboratory exercises: 26 hours

Educational activities Study phase: 10 hours

Language
This course is taught in Danish or English, depending on the lecturer. However, if international students participate, the teaching language will always be English.

Remarks
During the 3th quarter the course runs jointly with ST505/ST810 Statistical Simulation, and during the 4th quarter jointly with ST516/ST804 Computational Statistics (5 ECTS). Also runs together with the bachelor level course ST5XX Computational Statistics (10 ECTS).

Course enrollment
See deadline of enrolment.

Tuition fees for single courses
See fees for single courses.