MULTIVARIAT LINIER MIXED MODEL UNTUK PEMODELAN FAKTOR RESIKO KOMPLIKASI KEHAMILAN DAN PERSALINAN DENGAN PENDEKATAN MAXSIMUM LIKELIHOOD DI KABUPATEN TUBAN

DWI KURNIA PURNAMA SARI, 101614153018 (2018) MULTIVARIAT LINIER MIXED MODEL UNTUK PEMODELAN FAKTOR RESIKO KOMPLIKASI KEHAMILAN DAN PERSALINAN DENGAN PENDEKATAN MAXSIMUM LIKELIHOOD DI KABUPATEN TUBAN. Thesis thesis, Fakultas Kesehatan Masyarakat.

[img]
Preview
Text (abstrak)
abstrak.pdf

Download (127kB) | Preview
[img] Text (full text)
full text.pdf
Restricted to Registered users only until 11 October 2021.

Download (2MB)

Abstract

linear model is not just a classical/common linear model with a homogeneous variety that has one parameter and independent data, but many observations or cases are uncommon. A linear model with two or more different parameters is called a multivariate linear mixed model. Multivariate linear mixed model consists of two or more parameters, namely fixed factor and random effect. The purpose of this research was modeling of multivariate linear mixed model on risk factor of pregnancy and childbirth complication at Tuban District’s clinic. Specific objectives of this research were: (1) analyzing random effect and fixed effect variables; (2) formulating multivariate mixed linear models; (3) modeling estimates using multivariate linear mixed models; and (4) comparing the best modeling in risk factors for pregnancy and childbirth complications in Tuban District. The analysis results with software R showed mixed linear models on the number of maternal bleeding (Y1) and maternal post partum blood pressure (Y2). Predictor variables of fixed effect was blood pressure (X1), Hb levels (X2), nutritional status of pregnant mother (LILA) (X3), and pregnant mother's weight (X4). The predictor variables of random effect were obtained from the research clinic. With value -2(log likelihood) = 9844.772 > 2 431;0,05  480,403, then rejected the H0 so the regression parameters in the model were simultaneously significant with the linear equations: Y1 = 5,316 + 0,008X1 – 0,275x2 – 0,065X3 + 0,012 X4 – 0,055Z Y2= 134,86 – 0,169X1 – 5,766X2 + 1,176X3 + 0,865 X4 + 0,088Z The best model was the second equation with the largest likelihood log value of -2615.8264, while the likelihood first equation log value was -2614.8668. Therefore, the quality of antenatal care improvement needs to be considered so that the emergency can be detected early, especially on the variables that had been studied and the condition of pregnant mother comprehensively. Increased public awareness of the routine ANC importance and comprehensive health checks is also needed to detect pregnancy and childbirth complications.

Item Type: Thesis (Thesis)
Additional Information: KKC KK TKM 41/18 Sar m
Uncontrolled Keywords: multivariate linear mixed, obstetric complication, maximum likelihood estimation
Subjects: R Medicine > RA Public aspects of medicine > RA1-1270 Public aspects of medicine > RA421-790.95 Public health. Hygiene. Preventive medicine > RA428-428.5 Public health laboratories, institutes, etc.
Divisions: 10. Fakultas Kesehatan Masyarakat > Magister Ilmu Kesehatan Masyarakat
Creators:
CreatorsEmail
DWI KURNIA PURNAMA SARI, 101614153018UNSPECIFIED
Contributors:
ContributionNameEmail
ContributorKuntoro, Prof. , Dr., dr., MPH., PHUNSPECIFIED
Depositing User: Unnamed user with email indah.fatma@staf.unair.ac.id
Date Deposited: 11 Oct 2018 13:29
Last Modified: 11 Oct 2018 13:29
URI: http://repository.unair.ac.id/id/eprint/74604
Sosial Share:

Actions (login required)

View Item View Item