%0 Journal Article
%@ 978-979-96880-8-8
%A Warnida Lena, .
%A I Made Sumertajaya, .
%A Bagus Sartono, .
%D 2015
%F UNY:23644
%I Faculty of Mathematics and Sciences Yogyakarta State University
%J Proceeding of  International Conference On Research, Implementation And Education  Of Mathematics And Sciences 2015 (ICRIEMS 2015), Yogyakarta State University, 17-19 May 2015
%T THE METHODS OF VERTEX DISCRIMINANT MULTICATEGORY ANALYSIS WITH BOOTSTRAP APPLICATION
%U http://eprints.uny.ac.id/23644/
%X Underdeveloped district is the district that has less developed community and region compare with the other areas in the national scale base on  economical category, society, human resources, infrastructure, financial capacity, accessibility, and regional characteristics. It is not easy in classification of underdeveloped district that involves many variables and the number of observations by multicategory cases. Sometimes the data used does not fill the double normal assumption and the group of variance-covariance matrix is not a homogeneous. Verteks Discriminant Analysis (VDA) is the method of the newest  multicategories classification which can handle high-dimensional data. In this research, the significance testing of the function of vertex discriminant is done by using a bootstrap approach to determine the significantly variables. Through simultaneous confidence intervals of T^2-Hotelling, based on the results of analysis show that from the 27 predictor variables used, all significant variables have the real effect in determining the status of underdeveloped district. The accuracy of the classification that is obtained from the vertex discriminant function is about 94.44% for data training and 72.22% for data testing.  Keywords: underdeveloped district, VDA, bootstrap, simultaneous  confidence interval