PUTRI MEGASARI, 101041034 (2012) ANALISIS PARTIAL LEAST SQUARE REGRESSION MENGGUNAKAN DATA ANGKA KEMATIAN BAYI DI JAWA TIMUR TAHUN 2010. Thesis thesis, UNIVERSITAS AIRLANGGA.

Preview	Text (ABSTRAK) gdlhub-gdl-s2-2013-megasaripu-23107-8.abstr-t.pdf Download (176kB) | Preview
	Text (FULL TEXT) gdlhub-gdl-s2-2013-megasaripu-23107-23full.pdf Restricted to Registered users only Download (835kB) | Request a copy

Abstract

Infant Mortality Rate (IMR) is one of an important indicator for health status. The purpose of this study was to apply Partial Least Square Regression (PLSR) analysis for infant mortality rate model in East Java 2010. This is a non-reactive research, using secondary data without having direct interaction with the subject. The independent variable were birth trauma, asphyxia, low birth weight, infections and congenital abnormalities. The dependent variable was the Infant Mortality Rate (IMR). The conclusion of this study was PLSR models have higher accuracy than multiple regression models with backward stepwise method for IMR data in East Java 2010. The R² value was 17% in the PLSR models and 14.23% in multiple regression models with backward method. It is suggested that PLSR analysis had to be applied to an IMR model. Further studies on Logistik PLSR analysis shold be performed for the development of PLSR analysis.

Actions (login required)

View Item