delo diplomskega seminarja
Gašper Domen Romih (Author), Gašper Jaklič (Mentor), Sonja Bogatin (Co-mentor)

Abstract

Linearna regresija je ena izmed bolj uporabljenih statističnih metod, kar sledi iz dejstva, da se linearne modele lahko uporablja za veliko različnih problemov. V diplomski nalogi bo predstavljena večrazsežna linearna regresija in veljavnost večrazsežnega linearnega modela. Veljavnost modela lahko preverjamo z različnimi statističnimi testi samega modela in njegovih koeficientov. Poleg teoretičnega opisa bo v diplomski nalogi predstavljen primer uporabe večrazsežne linearne regresije. Na primeru "hydrostatic-season-time" modela, ki se uporablja za zagotavljanje varnosti na jezovih, bomo predstavili postopke izbire vplivnih spremenljivk, njihovo statistično ocenjevanje in oceno veljavnosti modela. Postopek izbire napovednih spremljivk je lahko precej zahteven in ne podaja enolično rešitev. Zato je potrebno v postopku izbire statistično vplivnih napovednih spremljivk in izbire optimalnega modela vedno opraviti več testov in opazovati različne parametre.

Keywords

matematika;hidrostatični-sezonsko-časovni model;večrazsežna linearna regresija;statistično vrednotenje;

Data

Language: Slovenian
Year of publishing:
Typology: 2.11 - Undergraduate Thesis
Organization: UL FGG - Faculty of Civil and Geodetic Engineering
Publisher: [G. D. Romih]
UDC: 519.2
COBISS: 18472281 Link will open in a new window
Views: 1181
Downloads: 424
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: Multiple linear regression analysis
Secondary abstract: Linear regression is one of the most used statistical techniques due to the fact that the linear model can be applied to many different problems. Multiple linear regression will be presented in diploma as well as the multiple linear regression model adequacy. The model adequacy can be assessed with different statistical tests of the model and its coefficients. Based on described theoretical backgrounds, the practical implementation of the multiple linear regression will be presented. Techniques of the parameter selection, their statistical evaluation and assessment of the model adequacy will be presented on the hydrostatic-season-time model, which is used to monitor dam activities and assuring it's safety. The parameter selection procedure can be very demanding and does not result an unique solution. Therefore it is necessary to perform various statistical tests and to observe several parameters to define statistically most influential parameters and an optimal model.
Secondary keywords: mathematics;hydrostatic-seasonal-time model;statistical valuation;multiple linear regression;
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, Matematika - 1. stopnja
Pages: 43 str.
ID: 10961364