Modeliranje pandemije COVID-a

master's thesis

• Title (english): COVID pandemic modeling
• Author: Nikolina Šarić
• Mentor: Pavle Goldstein (mentor)
• Committee member: Pavle Goldstein (predsjednik povjerenstva)
• Committee member: Miljenko Huzak (član povjerenstva)
• Committee member: Marko Tadić (član povjerenstva)
• Committee member: Marko Horvat (član povjerenstva)
• Granter: University of Zagreb; Faculty of Science; (Department of Mathematics); Zagreb
• Defense date and country: 2021-12-02, Croatia
• Scientific / art field, discipline and subdiscipline: NATURAL SCIENCES; Mathematics

Abstract

S obzirom na trenutnu situaciju, koronavirus trenutno je jedna od najčešćih tema novih istra živanja. U ovom radu ispitujemo i uspoređujemo dnevne stope zaraze susjednih zemalja i regija. Naime, razumno je očekivati da će stope zaraze zemalja ili regija koje su vrlo povezane utjecati jedna na drugu. Divergencija između stopa zaraze raznih država opažena je nakon 30.6.2021. vjerojatno zbog različitih postotaka procijepljenosti. Zbog toga podaci su podijeljeni na dva dijela: prije i poslije 30.6.2021. U prvom periodu dokazane su značajne korelacije izmedu svih odabranih država. U drugom periodu, značajna korelacija je nestala između nekih država. S koreliranim državama napravljeni su modeli linearne regresije te je utvrđena njihova zadovoljavajuća procjena. Da bi se isprobao drugačiji pristup podacima, diskretizirani podaci su na kraju pretvoreni u skrivljen Markovljev model.

Abstract (english)

Considering the current crisis, COVID-19 related topics are at the forefront of research. In this work, we are concerned with daily infection rates in neighbouring countries and regions. Namely, it stands to reason that infection rates in highly connected countries or regions will be related. However, divergence has been observed for the period after 30.6.2021., probably due to varying vaccination rates. For that reason, data was separated into two parts: before and after 30.6.2021. In the first period, significant correlation between all the countries was obtained. In the second period, significant correlation did not exist for some countries. Using the correlated countries, linear regression models were made and tested. To try a different approach to data analysis, discrete data was modeled into a hidden Markov model.

• Keywords: koronavirus; postotak procijepljenosti; modeli linearne regresije; skrivljen Markovljev model
• Keywords (english): coronavirus; vaccination rate; linear regression models; a curved Markov model
• Language: croatian
• URN:NBN: urn:nbn:hr:217:263394
• 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: 2021-12-16 13:37:05