Suharni, - (2015) Estimasi Model Probit Data Panel Pada Rata-Rata Jumlah Anak Lahir Hidup Di Provinsi Jawa Timur. Masters thesis, Institut Technology Sepuluh Nopember.
Preview |
Text
1312201020-Undergraduate Thesis.pdf - Published Version Download (1MB) | Preview |
Abstract
Salah satu model regresi dengan variabel respon kualitatif adalah regresi probit.
Regresi probit merupakan suatu model regresi yang berhubungan dengan unit-unit
probabilitas yang fungsi linknya menggunakan distribusi normal standar.
Pemodelan probit ini tidak hanya menggunakan data cross-section, tetapi dapat
dikembangkan menggunakan data panel yaitu gabungan antara data cross-section
dan data time series. Salah satu pendekatan model panel yang digunakan adalah
random effect dengan menggunakan Maximum Likelihood Estimation (MLE).
Estimasi merupakan bentuk integral rangkap yang sulit diselesaikan dengan cara
analitik sehingga digunakan metode kuadratur yaitu metode kuadrat Gauss
Hermite. Pengujian parameter yang digunakan dalam model probit data panel
adalah uji Likelihood Ratio dan uji Wald. Selanjutnya hasil kajian teoritis
diaplikasikan pada data real untuk mengetahui faktor-faktor yang mempengaruhi
rata-rata jumlah Anak Lahir Hidup (ALH) di provinsi Jawa Timur. Variabel yang
berpengaruh signifikan yaitu umur kawin pertama usia di bawah 17 tahun, Angka
Kematian Bayi (AKB) per 1000 kelahiran hidup, rata-rata lama sekolah dan laju
pertumbuhan ekonomi. Diperoleh probabilitas kabupaten/kota yang memiliki ratarata
jumlah ALH di atas rata-rata provinsi dan di bawah rata-rata provinsi dengan
ketepatan klasifikasi 70,6%.
=====================================================================================================
One of the regression model with qualitative response variable is the probit
regression. Probit regression is a regression model relating to units using the link
function of the probability that a standard normal distribution. Probit modeling is
not only using cross-section data, but can be developed using panel data which
combines cross-section data and time series data. One approach used is a panel
model with random effect using Maximum Likelihood Estimation (MLE).
Estimation is a form of double integral that difficult to solve with analytically so
used method Gauss Hermite quadrature. The test parameters used in the model is
the panel data probit Likelihood Ratio test and the Wald test. Furthermore, the
results of theoretical studies applied to real data to determine the factors that affect
the average number of Children Ever Born in the province of East Java. The
variables that have a significant effect of age at first marriage is under 17 years of
age, the Infant Mortality Rate (IMR) per 1,000 live births, the average length of
the school and the rate of economic growth. Retrieved probability district/city has
an average number of Children Born Alive above the provincial average and
below the provincial average with 70.6% classification accuracy.
| Item Type: | Thesis (Masters) |
|---|---|
| Additional Information: | RTSt 519.535 4 Suh e |
| Uncontrolled Keywords: | Gauss-Hermite, Maximum likelihood Estimation (MLE), Probit Data Panel, Random Effect, Rata-rata Anak Lahir Hidup |
| Subjects: | Q Science > QA Mathematics > QA278.5 Principal components analysis. Factor analysis. Correspondence analysis (Statistics) |
| Divisions: | Faculty of Mathematics and Science > Statistics > 49101-(S2) Master Thesis |
| Depositing User: | Mr. Tondo Indra Nyata |
| Date Deposited: | 01 Nov 2018 03:56 |
| Last Modified: | 01 Nov 2018 03:56 |
| URI: | http://repository.its.ac.id/id/eprint/59930 |
Actions (login required)
 |
View Item |