KOMPARASI METODE POHON REGRESI DENGAN METODE REGRESI COX UNTUK ANALISIS DATA SURVIVAL (ANALISIS FAKTOR YANG MEMPENGARUHI LAMA PERAWATAN PENYAKIT DIARE PADA BALITA DI RUMAH SAKIT)

NIZLEL HUDA, 090114305 M (2003) KOMPARASI METODE POHON REGRESI DENGAN METODE REGRESI COX UNTUK ANALISIS DATA SURVIVAL (ANALISIS FAKTOR YANG MEMPENGARUHI LAMA PERAWATAN PENYAKIT DIARE PADA BALITA DI RUMAH SAKIT). Thesis thesis, UNIVERSITAS AIRLANGGA.

[img]
Preview
Text (ABSTRAK)
gdlhub-gdl-s2-2006-hudanizlel-707-tkm_03-04.pdf

Download (1MB) | Preview
[img]
Preview
Text (FULL TEXT)
gdlhub-gdl-s2-2006-hudanizlel-707-tkm_03-04.pdf

Download (2MB) | Preview
Official URL: http://lib.unair.ac.id

Abstract

Regression analysis is one statistical tests commonly used for multifactorial data analysis. Regression analysis is interesting to use theoretically as it is based on mathematics. In addition, this analysis can be employed to investigate and model relations between variables. Regression analysis can be applied in variousdisciplines, such as engineering, physics, economics, management, biological and social sciences. Problem that most commonly found in health data is that they do not fulfill the classical assumptions. Such problem may be overcome by means of regression analysis that is not bound to an_v assumptions, such as regression trees method and Cox regression analysis. This study was aimed to survival data analysis (analyze factors affecting the length of hospitalization of underwifes with diarrhea) using regression trees method and Cox regression method. Data used in this study were those on the length of hospitalization of underfives with diarrhea in Dr Soetomo Hospital. The number of data in this study was 586, 57 of which were censored data, and the remaining 529 uncensored. The cross validated relative error was 0.922 ±0.041, the resubstitution relative error was 0.696 and the complexity was 45.497, the independent variables affecting the length of hospitalization in underfives with diarrhea, as revealed by means of regression trees method, were length of diarrhea, accompanying disseases, nutritional status, and dehydration, while variables affecting the length of hospitalization in underfives with diarrhea, as indicated by means of Cox regression method of a = 0.05, were length of diarrhea, accompanying diseases, and nutritional status.

Item Type: Thesis (Thesis)
Additional Information: KKC KK TKM 03/04 Hud k
Uncontrolled Keywords: Survival data analysis, regression trees method, Cox regression method
Subjects: Q Science > QA Mathematics > QA299.6-433 Analysis
R Medicine > RA Public aspects of medicine > RA1-1270 Public aspects of medicine > RA960-1000.5 Medical centers. Hospitals. Dispensaries. Clinics
R Medicine > RJ Pediatrics > RJ101 Child Health. Child health services
Divisions: 09. Sekolah Pasca Sarjana > Ilmu Kesehatan
Creators:
CreatorsNIM/NIDN
NIZLEL HUDA, 090114305 MUNSPECIFIED
Contributors:
ContributionNameNIDN/NIDK/NUP
ContributorKuntoro, Prof.,H.,dr.,MPH.,Dr.PHUNSPECIFIED
ContributorHari Basuki N., Dr.,dr.,M.KesUNSPECIFIED
Depositing User: Nn Anisa Septiyo Ningtias
Date Deposited: 2016
Last Modified: 10 Aug 2016 06:57
URI: http://repository.unair.ac.id/id/eprint/35784
Sosial Share:

Actions (login required)

View Item View Item