DoctorateOpen Access

Forecast modeling of environmental protection expenditures in the context of global public goods with artificial neural network and alternative methods: An application for metropolitan municipalities in Turkey

2023
0 views
0 downloads
Advisor: Doç. Dr. Ceyda Kükrer Mutlu ; Prof. Dr. Murat Çuhadar

Abstract (EN)

The aim of this study is to forecast the environmental protection expenditures of the metropolitan municipalities in Turkey for the period 2022: 4-2024: 3. With this forecasting, it is aimed to facilitate policy makers, especially local governments, in predicting how much share will be allocated in the budget for environmental protection services in the future, and about the funding source needed for action plans to be made for environmental protection. In this study, the environmental protection expenditures of metropolitan municipalities in the period 2007: 1 – 2022: 3 were modeled according to the "Holt-Winters' Seasonal Exponential Smoothing", " Multiplicative-Seasonal Box-Jenkins " and "Artificial Neural Networks" methods used in the time series. When the actual expenditures and the results produced by the models were compared, it has been determined that the artificial neural network is the method with the least of the mean absolute percentage error. As a result of many trials, it has been seen that the 11-delayed 12-neuron NAR (Nonlinear Autoregressive) model, created using the artificial neural network method, provides the highest estimation accuracy for real values. With the help of the developed model, the environmental protection expenditures were forecasted of the metropolitan municipalities was realized for the period 2022: 4 - 2024: 3. According to the results of the study, it is forecasted that the environmental protection expenditures of the metropolitan municipalities in Turkey for the next eight periods will be in an increasing trend in general and will be 5.17 Billion TL as of the third quarter of 2024. Taking this amount into account in the shares in the general budget tax revenues to be transferred from the central government to the metropolitan municipalities and developing a separate environmental financing strategy for the environmental protection services of the metropolitan municipalities will prevent future financial imbalances and increase the success of environmental policies.

Author

Dr. Selma Aydoğan

How to Cite

Selma Aydoğan (Doctorate thesis). Forecast modeling of environmental protection expenditures in the context of global public goods with artificial neural network and alternative methods: An application for metropolitan municipalities in Turkey, 2023, Afyon Kocatepe University.

License

Tüm Hakları Saklıdır

This work is shared under the specified license terms.

More theses from Afyon Kocatepe University