Comparision of Kernel Functions in Geographically Weighted Regression Model: Suicide Data as an Application

The traumatic traces of suicide in a society and the emotional devastation due to these losses make it very important to determine the causes of suicide. In this study, the number of suicides data was used for Turkey’s 81 provinces in 2019.The effects of factors affecting suicide and spatial differe...

Full description

Saved in:
Bibliographic Details
Main Authors: Tuba Koç, Pelin Akın
Format: Article
Language:English
Published: Istanbul University Press 2021-12-01
Series:Acta Infologica
Subjects:
Online Access:https://cdn.istanbul.edu.tr/file/JTA6CLJ8T5/EF41433D5BEB4E41A9A0DECF2C47A5C0
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The traumatic traces of suicide in a society and the emotional devastation due to these losses make it very important to determine the causes of suicide. In this study, the number of suicides data was used for Turkey’s 81 provinces in 2019.The effects of factors affecting suicide and spatial differences on suicide were analyzed and predicted with geographically weighted regression models (GWR). GWR models were applied with different kernel functions, and the best GWR model was found with the bisquare kernel function. Factors affecting suicide numbers were established as human development index, proportion of internet users, and numbers of unemployment. When the results were examined, it was seen that the number of suicides in the provinces was affected by different factors. In addition, the 2019 suicide numbers and predicted values were mapped, and the results were found to be quite similar. The province with the highest number of suicides across the country was Istanbul.
ISSN:2602-3563