Moguće odrednice dohotka stanovnika lokalnih jedinica u RH

master's thesis

• Title (english): Possible determinants of income of residents of local units in the Republic of Croatia
• Author: Tina Krištofić
• Mentor: Katarina Ott (mentor)
• Committee member: Katarina Ott (predsjednik povjerenstva)
• Committee member: Hrvoje Šikić (član povjerenstva)
• Committee member: Damir Bakić (član povjerenstva)
• Committee member: Marcela Hanzer (član povjerenstva)
• Granter: University of Zagreb; Faculty of Science; (Department of Mathematics); Zagreb
• Defense date and country: 2023-03-03, Croatia
• Scientific / art field, discipline and subdiscipline: NATURAL SCIENCES; Mathematics

Abstract

Ovaj rad analizira utjecaj odabranih varijabli na prosječni dohodak po stanovniku u jedinicama lokalne i područne samouprave Republike Hrvatske. Analizira se devet varijabli: prosječni porez po stanovniku, stopa nezaposlenosti, postotak stanovništva sa završenim visokim obrazovanjem, broj stanovnika, postotak stanovništva mladeg od 14 godina, postotak stanovništva starijeg od 65 godina, pristup moru, broj kućanstava s pristupom poljoprivrednom zemljištu i gustoća naseljenosti te se provjerava njihov utjecaj na prosječni dohodak po stanovniku. Analiza se provodi zasebno za 428 općina, 128 gradova i 21 županiju. Ekonomski podaci (dohodak, porez i stopa nezaposlenosti) dobiveni su za 2019. godinu, dok su se za demografske varijable koristio popis stanovništva iz 2011. Analiza podataka temelji se na generaliziranom linearnom modelu. Na tri su se modela putem programskoj jezika R dobile procjene utjecaja varijabli na dohodak po stanovniku na svakoj skupini podataka. Prvo se obradio model u kojem se nalaze svaka od izabranih varijabli zasebno, potom model koji je sadržavao sve varijable. Posljednji je model restringirana verzija punog modela dobiven analizom značajnosti varijabli u modelu i visine korelacije između tih varijabli. Puni i restringirani modeli usporedili su se pomoću statistike omjera vjerodostojnosti i u svakom skupu podataka zaključilo se da postoji model s manje varijabli koji jednako dobro ili bolje opisuje podatke od punog modela. Dobiveni rezultati ukazuju da prosječni porez po stanovniku utječe pozitivno na prosječni dohodak po stanovniku, ali s različitom statističkom značajnošću u različitim teritorijalnim jedinicama. Značajan utjecaj imaju stopa nezaposlenosti, koja utječe negativno, i postotak stanovništva sa završenim visokim obrazovanjem, koji ima pozitivan utjecaj. Varijable broja stanovnika, postotka stanovništva mlađe od 14 su u većini modela bile statistički neznačajne. Postotak stanovništva starijeg od 65 ima konzistentno negativan utjecaj, ali zbog različite razine značajnosti nije dobiven jasan rezultat. Pristup moru i pristup poljoprivrednom zemljištu imali su različit utjecaj i statističku značajnost ovisno o odabranoj teritorijalnoj jedinici i modelu te nije moguće donijeti zaključak na razini države bez daljnje analize. Gustoća naseljenosti, koja je bila dostupna samo za županije, ima različit smjer utjecaja i različitu statističku značajnost te također nije jasno definiran njezin utjecaj.

Abstract (english)

This paper analyzes the impact of selected variables on the average income per capita (p.c.) in local and regional self-government units of the Republic of Croatia. Nine variables are analyzed: average tax p.c., unemployment rate, percentage of the population with comple- ted tertiary education, number of inhabitants, percentage of the population younger than 14, percentage of the population older than 65, access to the sea, number of households with access to agricultural land and population density. The analysis is carried out separately for 428 municipalities, 128 cities and 21 counties. Economic data (income, tax and unemployment rate) were obtained for the year 2019, while the 2011 census was used for demographic variables. The data analysis is based on a generalized linear model. Estimates of the impact of variables on income p.c. on each group of data were obtained using the R programming language. First, the model containing each of the selected variables was processed separately, and then the model that contained all variables was analysed. The last model is a restricted version of the full model obtained by analysing the significance of the variables in the model and the level of correlation between those variables. The full and restricted models were compared using the likelihood ratio statistic, and in each data set it was concluded that there was a model with fewer variables that described the data as well or even better than the full model. The obtained results indicate that the average tax p.c. has a positive effect on the average income p.c., but with different statistical significance in different territorial units. The unemployment rate, which has a negative impact, and the percentage of the population with completed tertiary education, which has a positive impact, have a significant influence. The number of inhabitants, the percentage of the population under the age of 14 were statistically insignificant in the majority of models. The percentage of the population older than 65 has a consistently negative impact, but due to different levels of significance, no clear result was obtained. The variables signifying access to the sea and access to agricultural land had different influence and statistical significance depending on the selected territorial unit and model, so it was not possible to draw a conclusion at state level without further analysis. Population density, which was available only for counties, has different direction of influence and statistical significance, and its impact is also not clearly defined.

• Keywords: prosječni porez po stanovniku; stopa nezaposlenosti; postotak stanovništva sa završenim visokim obrazovanjem; broj stanovnika; postotak stanovništva mladeg od 14 godina; postotak stanovništva starijeg od 65 godina; pristup moru; broj kućanstava s pristupom poljoprivrednom zemljištu; gustoća naseljenosti
• Keywords (english): average tax per inhabitant; unemployment rate; percentage of the population with completed higher education; population; percentage of the population under the age of 14; percentage of population older than 65 years; access to the sea; number of households with access to agricultural land; population density
• Language: croatian
• URN:NBN: urn:nbn:hr:217:741873
• 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: 2023-08-17 12:36:43