KLASIFIKASI KINERJA PERUSAHAAN DI INDONESIA DENGAN MENGGUNAKAN METODE WEIGHTED K NEAREST NEIGHBOR (Studi Kasus 436 Perusahaan Yang Terdaftar Di Bursa Efek Indonesia Tahun 2015)

UTAMI, CYNTIA SURYA (2017) KLASIFIKASI KINERJA PERUSAHAAN DI INDONESIA DENGAN MENGGUNAKAN METODE WEIGHTED K NEAREST NEIGHBOR (Studi Kasus 436 Perusahaan Yang Terdaftar Di Bursa Efek Indonesia Tahun 2015). Undergraduate thesis, Fakultas Sains dan Matematika, Undip.

[img]
Preview
PDF
721Kb

Abstract

A company's performance can be seen from the analysis of the company's financial statements. Financial statement analysis is used to determine the development of the company's financial condition. In analyzing the financial statements required financial ratios depicting the weight of the company's performance. This thesis aims to classify the performance of the company into two classifications, namely the company healthy and unhealthy companies as well as determine the level of accuracy. This final project using financial ratio data 436 companies listed in the Indonesia Stock Exchange in 2015 which has been audited and is divided into two parts of 349 training data and 87 test data. The method used is the weighted k nearest neighbor with a dependent variable is the performance of the company and six independent variables are financial ratios WCTA, ROA, TATO, DAR, LDAR and ROI. The results of this thesis show that the method of calculation of weighted k k nearest neighbor optimal done by trial and error. Provided that the optimal k at k = 3 for kernel inversion, epanechnikov and triangles while for optimal kernel k gauss at k = 4. The accuracy of classification and classification performance of the company gave almost the same results by using kernel inversion, Gauss, epanechnikov and triangles. Keywords: financial ratios, weighted k nearest neighbor and kernel inversion, Gauss, epanechnikov and triangles.

Item Type:Thesis (Undergraduate)
Subjects:H Social Sciences > HA Statistics
Divisions:Faculty of Science and Mathematics > Department of Statistics
ID Code:55074
Deposited By:Mr Hasbi Yasin
Deposited On:26 Jul 2017 10:49
Last Modified:26 Jul 2017 10:49

Repository Staff Only: item control page