Gayrisafi yurtiçi hasılanın coğrafi dağılımının çok boyutlu ölçekleme analizi kümeleme analizi ve temel bileşenler analizleri ile karşılaştırmalı olarak incelenmesi

The comparative investigation of the geographical research of the gross interior native land product with the multidimensional scaling,cluster analysis and principal component analysis

Özet

Çok boyutlu ölçekleme analizi (Multidimensional scaling, MDS). n nesne ya da birim arasındaki p değişkene göre belirlenen uzaklıklara dayalı olarak nesnelerin k boyutlu (k

Abstract

Multidimensional scaling. MDS, is a method that aims to obtain the indication of objects in a k-dimensional space on the basis of the dislances which defines according to the p variables between n objects or cases, so benefits to define the relations between the objects and apply in many fields. Medicine, psychiatry, social sciences, education sciences, market researches, etc. For example, in psychology in the cases that the similar or different individuals can not be create clearly, by the aim of defining the order according to their similarities of the similar facts, we profit from MDS. In our study, by the aim of defining the geographical distribution of the gross interior native land product, we used the MDS method. So far as it knows, geographical distribution of the gross interior native land product of a country is one of the most appropriate indicators which shows the development level ol the regions an the cities in that country. Because of this, in our study the gross interior native land product percentages of the sectors of agriculture, industry and service of seven geographical regions arc used. Right along with MDS, the cluster analysis and the principal component analysis are applied to the data too, and the results are interpreted contrastively.