NMAK15017U Inference in Hidden Markov Models

Volume 2015/2016
Education

MSc programme in Statistics
 

Content

Hidden Markov models: Definition and properties. Estimation by direct maximization of the likelihood, by the EM algorithm and further MC methods. Forecasting, decoding and prediction.

 

Learning Outcome

Knowledge:
The student should know what Hidden markov models are and know about different inference methods.
Skills:
The students shall be able to set up hidden Markov models, and obtain insight into accessible methods of parameter estimation, and apply them to relevant models.
Competences:
The student should be able to generalize from the specific models introduced in the course to specific problems encountered further on.

Stok and Stat 1+2
2 + 2 hours of lectures and 2 hours of exercise sessions per week for 7 weeks.
  • Category
  • Hours
  • Exam
  • 27
  • Lectures
  • 28
  • Practical exercises
  • 7
  • Preparation
  • 137
  • Theory exercises
  • 7
  • Total
  • 206
Credit
7,5 ECTS
Type of assessment
Written assignment, 27
Written take-home assignment (handing-out at 9 am and submission at 12 noon the following day).
Exam registration requirements

To register for the exam, the student has to present exercises in class at one of the exercise sessions.

Aid
All aids allowed
Marking scale
7-point grading scale
Censorship form
No external censorship
Re-exam

Same as ordinary exam. If the requirement of presenting exercises in class during the course is not fulfilled, a written assignment is due two weeks before the re-exam. The assignment will be handed out three weeks before the re-exam.

Criteria for exam assesment

The student must in a satisfactory way demonstrate that he/she has mastered the learning outcome of the course.