Analiza COVID-19 podataka analizom panel podataka

master's thesis

• Title (english): Analysis of COVID-19 data using panel data analysis
• Author: Mia Gojević
• Mentor: Anamarija Jazbec (mentor)
• Committee member: Goran Muić (predsjednik povjerenstva)
• Committee member: Anamarija Jazbec (član povjerenstva)
• Committee member: Pavle Goldstein (član povjerenstva)
• Committee member: Hrvoje Planinić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Science; (Department of Mathematics); Zagreb
• Defense date and country: 2022-07-20, Croatia
• Scientific / art field, discipline and subdiscipline: NATURAL SCIENCES; Mathematics

Abstract

Ovaj rad baziran je na analizi statističkog i ekonometrijskog pojma naziva panel podaci. Analiza panel podataka vrlo je raširena u prirodoslovnim, društvenim i ekonomskim znanstvenim istraživanjima zbog mogućnosti praćenja više objekata putem više varijabli u više od jednog vremenskog perioda. U prvom poglavlju definirani su panel podaci kao kombinacija vremenskih nizova i vremenskih presjeka te je dana njihova struktura. Kako bi bilo lakše razumijevanje linearnih modela s takvim podacima, za početak je objašnjena jednostavna linearna regresija te statistički pojmovi potrebni za analizu. Potom je prikazana osnovna konstrukcija linearnih modela za panel analizu te dana teorijska podloga za združeni model, model fiksnih efekata i model slučajnih efekata. Za odabir najboljeg modela od navedenih primijenjeni su F-test, Hausmann test i Lagrange Multiplier Test - (Breusch-Pagan). U drugom poglavlju provedena je analiza u programu za statističku analizu R nad panel podacima o COVID-19 bolesti u Hrvatskoj, Italiji, Švicarskoj i švedskoj. U vremenskom razdoblju od svibnja 2020. godine do studenog 2021. godine promatrana je zavisna varijabla broj smrtnih slučajeva s obzirom na nezavisne varijable: broj zaraženih osoba, broj hospitaliziranih osoba, broj cijepljenih osoba barem jednom dozom cjepiva, indeks strogosti mjera te indeks ljudskog razvoja.

Abstract (english)

Fundamental part of this paper is analysis of statistical and econometrical term panel data. Panel analysis is used for scientific researches in fields like economy and nature and social sciences. It is very popular because it can handle multiple objects over multiple variables within more than one time period. In first chapter, it is shown structure and definition of panel data as a combination of time series and cross-sectional data. Also, there is given an explanation of a simple linear regression and main statistical tools for validating linear models. Afterwards, basic structure of a linear panel model, pooled OLS model, fixed effects model and random model effects are presented. To determine which model fits best with the panel data there are conducted F-test, Hausman test and Lagrange Multiplier Test - (Breusch-Pagan). Second chapter gives overview of panel data analysis through example made in program for statistical analysis R. Panel is consisted of COVID-19 data from Croatia, Italy, Sweden and Switzerland from May 2020. until November 2021. Dependent variable is number of death cases opposite to independent variables that are number of total positive cases, number of patients in hospital, number of vaccinated patients with at least 1 dose, stringency index and human development index.

• Keywords: statistički i ekonometrijski pojam; vremenski nizovi; vremenski presjeci; linearni modeli; linearna regresija; model fiksnih efekata; model slučajnih efekata; F-test; Hausmann test; Lagrange Multiplier Test - (Breusch-Pagan); COVID-19 bolest
• Keywords (english): statistical and econometrical term; time series; cross-sectional data; linear models; linear regression; fixed effects model; random effects model; F-test; Hausmann test; Lagrange Multiplier Test - (Breusch-Pagan); COVID-19 disease
• Language: croatian
• URN:NBN: urn:nbn:hr:217:318596
• 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
• Created on: 2022-10-26 10:10:11