Kompleksnost skrivenih Markovljevih modela

Mišura, Ankica (2016) Kompleksnost skrivenih Markovljevih modela. Diploma thesis, Faculty of Science > Department of Mathematics.

[img]
Preview
PDF
Language: Croatian

Download (1MB) | Preview
[img] Archive
Language: Croatian

Download (18MB)

Abstract

This thesis is concerned with a statistical model called hidden Markov model (HMM), statistical tool designed for modelling sequences generated by hidden processes. We give a formal definition of the hidden Markov model, describe several algorithms used in their analysis and present their in Python. We also construct example of hidden Markov model. We simulate data and attempt to find the best model for it. We use several statistical methods: likelihood maximization, log-likelihood ratio and information criteria.

Item Type: Thesis (Diploma thesis)
Supervisor: Goldstein, Pavle
Date: 2016
Number of Pages: 32
Subjects: NATURAL SCIENCES > Mathematics
Divisions: Faculty of Science > Department of Mathematics
Depositing User: Iva Prah
Date Deposited: 03 Nov 2016 13:12
Last Modified: 03 Nov 2016 13:12
URI: http://digre.pmf.unizg.hr/id/eprint/5259

Actions (login required)

View Item View Item