delo diplomskega seminarja
Jaka Munda (Author), Aljoša Peperko (Mentor)

Abstract

V diplomski nalogi obravnavamo primera podatkov z manjkajočimi podatki in primer brez manjkajočih podatkov, ki izhajajo iz zaporedja slučajnih vektorjev, ki so neodvisno enako porazdeljeni z večrazsežno normalno porazdelitvijo s parametroma vektorjem matematičnega upanja in kovariančno matriko. Za vsako obliko podatkov lahko po metodi največjega verjetja izračunamo cenilki parametrov porazdelitve. Pristopov za izračun cenilke po metodi največjega verjetja je več, v delu obravnavamo pristopa z matričnim odvajanjem in matrično transformacijo. Obravnavamo še monoton vzorec, ki je poseben primer manjkajočih podatkov, za katerega prav tako izračunamo cenilki za parametra po metodi največjega verjetja.

Keywords

matematika;večrazsežna normalna porazdelitev;metoda največjega verjetja;matrično odvajanje;monotoni vzorci;

Data

Language: Slovenian
Year of publishing:
Typology: 2.11 - Undergraduate Thesis
Organization: UL FS - Faculty of Mechanical Engineering
Publisher: [J. Munda]
UDC: 519.2
COBISS: 18737497 Link will open in a new window
Views: 1593
Downloads: 233
Average score: 0 (0 votes)
Metadata: JSON JSON-RDF JSON-LD TURTLE N-TRIPLES XML RDFA MICRODATA DC-XML DC-RDF RDF

Other data

Secondary language: English
Secondary title: Maximum likelihood estimation of the parameters of a multivariate normal distribution
Secondary abstract: In this paper we consider sample with missing data and sample without missing data, that comes from multivariate normal distribution with parameters mean vector and covariance matrix. No matter the shape of the data we can estimate parameters with maximum likelihood estimation. There are various techniques for estimating parameters with maximum likelihood estimation. We consider two techniques, namely, matrix differentiation and matrix transformation. With both techniques we must derivate likelihood function that we get from the sample. We also consider monotone sample, which is a special case of missing data for which we can also estimate parameters with method of maximum likelihood estimation.
Secondary keywords: mathematics;multivariate normal distribution;maximum likelihood estimation;matrix differentiation;monotone sample;
Type (COBISS): Final seminar paper
Study programme: 0
Embargo end date (OpenAIRE): 1970-01-01
Thesis comment: Univ. v Ljubljani, Fak. za matematiko in fiziko, Oddelek za matematiko, Finančna matematika - 1. stopnja
Pages: 28 str.
ID: 11229756
Recommended works:
, delo diplomskega seminarja
, delo diplomskega seminarja
, delo diplomskega seminarja