The examination of provinces in Turkey in terms of livestock statistics variable with multivariate statistical methods

In this study, multivariate statistical techniques such as clustering analysis, factor analysis and multidimensional scaling analysis are given theoretically. In addition, animal existence containing 29 variables of the livestock statistics obtained from TÜİK in 2018 and animal production data are examined with multivariate analysis techniques. Firstly, in the study clustering analysis is included, the most appropriate cluster structure is obtained by the Ward method which is one of the hierarchical clustering methods and, the provinces examined in this study are divided into two groups. The number of the suitable clusters in non-hierarchical clustering analysis is determined by discriminant analysis by using cluster parameters as well as some parameters, moreover; Fuzzy clustering method is concluded to be the most effective method in the cluster structure of 5. Furthermore, the per capita animal production, the raw data of the animal existence, and livestock development indices of the provinces with standardized data are calculated by using factor loadings obtained from factor analysis in the study. It is seen that Ardahan is the first province whereas İstanbul is the last province in both rankings. Finally, metric multidimensional scaling technique was used.