Analisis Diskriminan dalam Menentukan Fungsi Pengelompokan Kabupaten/Kota di Indonesia berdasarkan Indikator Indeks Pembangunan Manusia

Nurhasanah Nurhasanah
Nany Salwa (Program Studi Statistika, Fakultas Matematika dan Ilmu Pengetahuan Alam, Universitas Syiah Kuala)
Lyra Ornila (Program Studi Statistika, Fakultas Matematika dan Ilmu Pengetahuan Alam, Universitas Syiah Kuala)
Fitriana AR (Program Studi Statistika, Fakultas Matematika dan Ilmu Pengetahuan Alam, Universitas Syiah Kuala)
Amiruddin Hasan (Program Studi Pendidikan Ekonomi, Fakultas Keguruan dan Ilmu Pendidikan, Universitas Syiah Kuala)

Jurnal EMT KITA
Vol 5 No 1 (2021): JANUARY-JUNE 2021

The Human Development Index (HDI) is a measure used to measure the success of human development in an area. There are several indicators used to compile the HDI value. Previously, regencies/cities were grouped based on the HDI indicator. The grouping is done using K-means cluster analysis with 4 categories, namely regencies/cities that have low, medium, high, and very high HDI indicator values. From the results of determining the category of the HDI indicator in an area, we need a function that can be used to classify an object into one of the HDI indicator value categories. The compilation of the classification function is carried out using discriminant analysis. The results obtained from the discriminant analysis are that there are 10 variables or indicators that fall into the discriminant function. The resulting discriminant function is quite good in classifying each group with a success rate of more than 85% and the discriminant function is supported by a fairly good validation success rate, namely with a classification accuracy of 93.20%.