Hipotēžu pārbaude divu izlašu lokācijas-mēroga modelim
Loading...
Date
Authors
Advisor
Journal Title
Journal ISSN
Volume Title
Publisher
Latvijas Universitāte
Language
lav
Abstract
Darbā aplūkots lokācijas-mēroga modelis un metodes tā noteikšanai. Galvenokārt tika aplūkoti lokācijas-mēroga testi, kā arī Kolmogorova-Smirnova tests, un to robežsadalījumu uzvedība situācijās, kad dati tika transformēti, zinot parametru nobīdi, kā arī izmantojot Mallova distances metodi. Lokācijas-mēroga modeļa apskatei tika pielietoti arī PP, QQ gafiki, un empīriskās ticamības funkcijas metode. Darba gaitā tika pētīta testu uzvedība, testējot gan simulētu, gan reālu datu piemērus.
This thesis contains an overview of the location-scale model and methods for its determination. The focus was on location-scale tests, as well as the Kolmogorov-Smirnov test, and the behavior of their limiting distributions in situations where data was transformed by knowing the parameter shift, as well as using the Mallows distance method. PP, QQ graphs, and the empirical likelihood function method were also used to inspect the location-scale model. During this thesis, the test behavior was studied by testing examples of both simulated and real data.
This thesis contains an overview of the location-scale model and methods for its determination. The focus was on location-scale tests, as well as the Kolmogorov-Smirnov test, and the behavior of their limiting distributions in situations where data was transformed by knowing the parameter shift, as well as using the Mallows distance method. PP, QQ graphs, and the empirical likelihood function method were also used to inspect the location-scale model. During this thesis, the test behavior was studied by testing examples of both simulated and real data.