PENERAPAN ANALISIS FAKTOR METODE KLASIK DAN METODE ROBUST MINIMUM COVARIANCE DETERMINANT

Silmi Firdausi Mahfudz, NIM. 14610007 (2018) PENERAPAN ANALISIS FAKTOR METODE KLASIK DAN METODE ROBUST MINIMUM COVARIANCE DETERMINANT. Skripsi thesis, UIN SUNAN KALIJAGA YOGYAKARTA.

Preview	Text (PENERAPAN ANALISIS FAKTOR METODE KLASIK DAN METODE ROBUST MINIMUM COVARIANCE DETERMINANT) 14610007_BAB-1_ VI _DAFTAR-PUSTAKA.pdf - Published Version Download (5MB) | Preview
	Text (PENERAPAN ANALISIS FAKTOR METODE KLASIK DAN METODE ROBUST MINIMUM COVARIANCE DETERMINANT) 14610007_BAB II_sampai_BAB V.pdf - Published Version Restricted to Registered users only Download (9MB)

Abstract

Factor analysis is a statistical method used to describe a set of variables based on common dimensions. Factor analysis that is often used is classical factor analysis with principal component method (PCA) .PCA uses a covariant variant matrix that can not work properly if the analyzed data contains outliers. In this research PCA was developed by estimating the covariance variant matrix with a robust estimator Minimum Covariance Determinant (MCD). Estimator MCD is one of the robust estimator that aims to get the smallest determinant of the covariance matrix. Through factor analysis is expected to obtain robust high accuracy analysis result for data countaining many outlier. Both methods in this final assignment are applied to explain the number of foodstuff subgroups in the 2016 average data on consumption expenditure in the Special Region of Yogyakarta into a simpler dimension. The percentage of cumulative diversity that can be explained by factors that are formed through factor analysis of classical methods on consumption expenditure data according to foodstuffs group in DIY is 54,5% lower than the robust factor analysis method that produces 58,2%. This shows that the robust factor analysis in this case is more efficient than classical factor analysis.

Share this knowledge with your friends :

Actions (login required)

View Item