Primjena statističkih metoda na analizu utjecaja demografskih varijabli u maloprodaji

master's thesis

• Author: Martina Škrnjug
• Mentor: Siniša Slijepčević (mentor)
• Committee member: Siniša Slijepčević (predsjednik povjerenstva)
• Committee member: Ilja Gogić (član povjerenstva)
• Committee member: Boris Guljaš (član povjerenstva)
• Committee member: Igor Pažanin (član povjerenstva)
• Granter: University of Zagreb; Faculty of Science; (Department of Mathematics); Zagreb
• Defense date and country: 2018-03-02, Croatia
• Scientific / art field, discipline and subdiscipline: NATURAL SCIENCES; Mathematics

Abstract

Prvi cilj ovoga rada je teorijski predstaviti model višestruke linearne regresije kao i metode odabira podskupa i metode sažimanja koje se koriste za njegovo poboljšanje. Kako bismo to napravili, rad smo započeli s teorijskom obradom modela jednostruke linearne regresije. Drugi je cilj pokušati statistički opisati vezu između godišnje prodaje određenog proizvoda i demografskih čimbenika. U tu smo svrhu koristili metodu višestruke linearne regresije koju smo prethodno teorijski obradili. Podaci su prvo prilagođeni, a nakon toga je provedena opisna statistika, modeliranje podataka i analiza modela. Na kraju rada iskazan je zaključak o optimalnosti korištenja modela linearne regresije za dane podatke.

Abstract (english)

The first goal of this paper is to theoretically present the multiple linear regression model as well as the subset selection methods and the shrinkage methods that are used for its improvement. In order to do so, we have started the paper with theoretical analysis of a univariate linear regression model. The second goal is to try to describe the relationship between annual sales of a particular product and demographic factors. For this purpose, we used the method of multiple liner regression which we have previously theoretically presented. Data was first adjusted, and then descriptive statistics, data modeling, and model analysis were performed. At the end of the paper, a conclusion was reached on the optimum utilization of linear regression model for given data.

• Keywords: linearna regresija; višestruka linearna regresija; metode odabira podskupa; metode sažimanja
• Keywords (english): linear regression; multiple linear regression; subset selection methods; shrinkage methods
• Language: croatian
• URN:NBN: urn:nbn:hr:217:180926
• Study programme: Title: Finance and Business Mathematics; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra matematike (magistar/magistra matematike)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Open access
• Created on: 2018-11-20 13:11:44