Analiza prodaje motora višefaktorskom analizom varijance i analizom korespondencije

master's thesis

• Author: Anđela Iljadica-Rapo
• Mentor: Anamarija Jazbec (mentor)
• Committee member: Anamarija Jazbec (predsjednik povjerenstva)
• Committee member: Miljenko Huzak (član povjerenstva)
• Committee member: Boris Muha (član povjerenstva)
• Committee member: Marko Vrdoljak (član povjerenstva)
• Granter: University of Zagreb; Faculty of Science; (Department of Mathematics); Zagreb
• Defense date and country: 2017-09-27, Croatia
• Scientific / art field, discipline and subdiscipline: NATURAL SCIENCES; Mathematics

Abstract

U ovom radu opisana je analiza prodaje motora firme iz Šibenika koja se uspjela održati za vrijeme "prve globalne financijske krize 21. stoljeća", kad su mnogi prodavatelji motora u RH "propali". Podaci s kojima se radi sežu od 2006. do 2016. godine te zapravo opisuju veliki pad u prodaji. Poznate su jačine, boje, marke i tipovi prodanih motora u navedenom razdoblju. Rad se sastoji od 5 poglavlja u kojima su obrađene različite statističke metode. Prvo se opisnom statistikom opisala baza podataka, navedene su tablice frekvencija i mozaik prikazi varijabli baze. Potom se uradio \(\chi^2\)-test, jedan od prvih statističkih testova. Glavna tema trećeg poglavlja je analiza varijance. Nadalje, opisana je novija, jako fleksibilna statistička metoda, analiza korespondencije. U posljednjem poglavlju je obrađena linearna regresija kojom se izračunalo koliki pad u prodaji se dogodio svakoj od marki motora. Također, pokušala se predvidjeti daljnja prodaja. Prilikom obrade uzetih podataka korišten je programski sustav SAS, te su navedeni potrebni kodovi. Iako se radi o velikom padu u prodaji motora, iz tablica frekvencija se može vidjeti lagani porast u prodaji prošle godine (2016.), a djelatnici firme kažu da je porast još i veći ove godine.

Abstract (english)

This paper describes motor sales analysis of a company from Šibenik which survived during ”the first global financial crisis of the 21st century”, when most of Croatian motor dealers collapsed. The data extend from 2006 to 2016 and actually, sales come down is showed. Strengths, colors, brands and types of sold motors in the specified period are known. The paper consists of five chapters which show different statistic methods. First, the data base is described with descriptive statistics, and also frequency tables and mosaic plots of the base variables are showed. Secondly, \(\chi^2\)-test, one of the first statistic tests is done. The main theme of the third chapter was variance analysis. Then is described the newer, very flexible statistic method, correspondence analysis. The last chapter describes linear regression, which is used to calculate the exact sales come down for all of the motor brands. Also, we tried to make a prediction of the future sales. For all data analysis program system SAS is used and there are all required codes. Although the paper is about the big sales collapse, from the frequency table we can see light increase in sales from the last year (2016), and the company employees state that sale increase is even greater this year.

• Keywords: analiza varijance; analiza korespondencije; linearna regresija; SAS
• Keywords (english): variance analysis; correspondence analysis; linear regression; SAS
• Language: croatian
• URN:NBN: urn:nbn:hr:217:066015
• Study programme: Title: Mathematical Statistics; 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
• Repository: Repository of the Faculty of Science
• Created on: 2018-01-30 14:11:20