Hanehalkı tüketim harcamalarına göre İibs Düzey 2 bölgelerin kümelenmesi
Loading...
Date
2022
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Aksaray Üniversitesi
Access Rights
info:eu-repo/semantics/openAccess
Abstract
Mevcut çalışmada, Türkiye’de, tüketim harcamalarına göre benzer bölgelerin belirlenmesi amaçlanmıştır. Bu bağlamda, Türkiye İstatistik Kurumu’nun (TÜİK) yayınlamış olduğu Hanehalkı Tüketim Harcamasının verileri kullanılmıştır. İlgili veride gıda ve alkolsüz içecekler, alkollü içecekler, sigara ve tütün, giyim ve ayakkabı, konut ve kira, mobilya, ev aletleri ve bakım hizmetleri, sağlık, ulaştırma, haberleşme, eğlence ve kültür, eğitim hizmetleri, lokanta ve oteller, çeşitli mal ve hizmetler olmak üzere toplamda 12 harcama mevcuttur. Ayrıca veride, TÜİK’in belirlediği Türkiye İstatistiki Bölge Birimleri Sınıflandırması Düzey 2 bölgeleri yer almaktadır. Benzer bölgelerin belirlenmesi için kümeleme yöntemlerinden k-ortalamalar ve k kırpılmış ortalamalar kullanılmıştır. K-ortalamalar yönteminde küme sayısının belirlenmesi için ise Silhouette İndeksinden yararlanılmıştır. Analiz sonucunda 3 küme olduğu tespit edilmiş ve sırasıyla kümelerde 4, 4 ve 18 bölge yer almıştır. K kırpılmış ortalamalar analiz sonucuna göre 3 küme olduğu tespit edilmiş ve sırasıyla kümelerde 4, 10 ve 11 bölge yer almıştır TR10 (İstanbul) aykırı gözlem olarak belirlenmiştir.
This paper attempts to determine similar regions in Turkey according to consumption expenditures. For this reason, the data used in the analysis was gathered from Turkish Statistical Instıtute, Household Consumption Expenditures research. In the data, there were 12 variables: food and non-alcoholic beverages, alcoholic beverages, cigarettes, clothing and foot wear, housing and rent, furniture, house appliances, health, transportation, communication, entertainment and culture, educational services, restaurants, food services and hotels, various goods and services. Moreover, NUTS Level 2 regions was analyzed. In order to determine similar regions, k-means and k-trimmed means from clustering methods were used. The Silhouette Index was used to determine the number of clusters in the k-means method. As a result of the analysis, it was determined that there were 3 clusters and there were 4, 4 and 18 regions in the clusters, respectively. According to the results of the trimmed k-means analysis, it was determined that there were 3 clusters and there were 4, 10 and 11 regions in the clusters, respectively. TR10 (Istanbul) was determined as outlier.
This paper attempts to determine similar regions in Turkey according to consumption expenditures. For this reason, the data used in the analysis was gathered from Turkish Statistical Instıtute, Household Consumption Expenditures research. In the data, there were 12 variables: food and non-alcoholic beverages, alcoholic beverages, cigarettes, clothing and foot wear, housing and rent, furniture, house appliances, health, transportation, communication, entertainment and culture, educational services, restaurants, food services and hotels, various goods and services. Moreover, NUTS Level 2 regions was analyzed. In order to determine similar regions, k-means and k-trimmed means from clustering methods were used. The Silhouette Index was used to determine the number of clusters in the k-means method. As a result of the analysis, it was determined that there were 3 clusters and there were 4, 4 and 18 regions in the clusters, respectively. According to the results of the trimmed k-means analysis, it was determined that there were 3 clusters and there were 4, 10 and 11 regions in the clusters, respectively. TR10 (Istanbul) was determined as outlier.
Description
Keywords
Hanehalkı Tüketim Harcaması, Kümeleme Analizi, K-ortalamalar, K Kırpılmış Ortalamalar, Household Consumption Expenditures, Cluster Analysis, K-means, Trimmed k-Means
Journal or Series
Aksaray Üniversitesi İktisadi ve İdari Bilimler Fakültesi Dergisi
WoS Q Value
Scopus Q Value
Volume
14
Issue
4
