Geostatistical Analysis of Fiber-Optic Cable Investments in Türkiye

Date

2024-4-25

Author

Mevsim Tok, Nesrin

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

116
views

177
downloads

Cite This

Information and communication technologies (ICT) and broadband internet are indispensable for the era we live in. In addition to facilitating the daily activities and access to information, the internet also increases the development level and economic wealth of a society and government. Fiber technologies as the most advanced way of connecting to the internet have significant contributions on providing enhanced internet experience for residential and business users by ensuring high-speed, low-latency and reliable connectivity. Moreover, every single person and organization are expected to have qualified internet access under similar conditions without being exposed to digital divide that can be ensured by the advantageous characteristics of fiber infrastructures. However, fiber infrastructures have high deployment and operational costs and hence, telecommunication operators are constrained to make rational investments. So that, it is important to investigate the motivations behind operators' investment decisions as well as finding out if there are spatial variations of investments. The aim of this study is to explore and analyse the factors affecting the fiber investments while taking into account the geographic aspects throughout the study area, which is determined as the NUTS-3 statistical regions (81 provinces) of Türkiye. With exploratory analysis, GDP per capita, GDP of information and communication activity, number of schools and new buildings given construction permit are determined as the significant factors influencing the fiber investments. One global regression (Ordinary Least Squares-OLS) and two local regression models (Geographically Weighted Regression-GWR and Multiscale Geographically Weighted Regression-MGWR) are conducted within the study. According to the model diagnostics, in comparison with OLS, both GWR and MGWR provide improved results by considering spatial variations. While MGWR model is superior to GWR model in terms of model fit criteria and parameter estimates, GWR model comes forward in addressing local multicollinearity. Spatial regression results are mapped in order to get local insights and reveal the variations among the 81 provinces. It is observed that GDP per capita, GDP of information and communication activity and new buildings given construction permit have higher impacts on fiber investments across the provinces in eastern part, whereas number of schools is more influent across western part of Türkiye. There are significant contributions of this study to the literature which are suggesting new response and explanatory variables, considering NUTS-3 statistical regions (81 provinces), enriching the analysis with spatial statistical techniques and presenting mapping insights.

Subject Keywords

Fiber Investment, Digital Divide, Spatial Analysis, GWR, MGWR

URI

https://hdl.handle.net/11511/109806

Collections

Graduate School of Natural and Applied Sciences, Thesis