Primjena statističkih metoda u optimizaciji marketinških aktivnosti u maloprodaji

master's thesis

• Author: Sara Markić
• Mentor: Siniša Slijepčević (mentor)
• Committee member: Siniša Slijepčević (predsjednik povjerenstva)
• Committee member: Sonja Štimac (član povjerenstva)
• Committee member: Dražen Adamović (član povjerenstva)
• Committee member: Tomislav Berić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Science; (Department of Mathematics); Zagreb
• Defense date and country: 2017-11-29, Croatia
• Scientific / art field, discipline and subdiscipline: NATURAL SCIENCES; Mathematics

Abstract

Marketinške aktivnosti u maloprodaji bitna su komponenta i veći dio troška marketinške strategije poduzeća. Vrlo je korisno imati sustave koji bi mogli pomoći pri procjeni učinkovitosti specifične marketinške aktivnosti. U ovom radu ostvaren je model za predviđanje ishoda (zarade) uzevši u obzir karakteristike marketinške aktivnosti. Dan je osvrt na dvije metode iz područja matematičke statistike i strojnog učenja. Opisan je model linearne regresije i aditivnih stabala te je za svaki opisana metoda učenja. Dana je i usporedba između više inačica ostvarenih modela te je naposljetku vrednovana korisnost najbolje inačice.

Abstract (english)

Marketing activities in retail are an important component and major cost in marketing strategy of a company. It is very useful to have systems which are capable of estimating the outcome of a specific marketing activity. This thesis describes the model for predicting the outcome (yield) of a marketing activity by taking all of its characteristics into account. Two methods from the field of mathematical statistics and machine learning are described. Linear regression and additive trees models are covered and for each a description of learning methods is given. Comparison between several implemented models is given and evaluation of usefulness is done for the best model.

• Keywords: marketinške aktivnosti; procjena učinkovitosti; model za predviđanje ishoda; model linearne regresije; model aditivnih stabala
• Keywords (english): marketing activities; estimating outcome; predicting yield; linear regression; additive trees
• Language: croatian
• URN:NBN: urn:nbn:hr:217:056173
• 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: 2018-05-28 09:16:50