Analisis regresi Cox Proportional Hazard dan Accelerated Failure Time pada data waktu tunggu kerja

Rayyan, Nizma Aini (2023) Analisis regresi Cox Proportional Hazard dan Accelerated Failure Time pada data waktu tunggu kerja. Sarjana thesis, UIN Sunan Gunung Djati Bandung.

Preview	Text (COVER) 1_cover.pdf Download (858kB) | Preview
Preview	Text (ABSTRAK) 2_abstrak.pdf Download (48kB) | Preview
Preview	Text (DAFTAR ISI) 3_daftarisi.pdf Download (93kB) | Preview
Preview	Text (BAB 1) 4_bab1.pdf Download (189kB) | Preview
	Text (BAB 2) 5_bab2.pdf Restricted to Registered users only Download (312kB) | Request a copy
	Text (BAB 3) 6_bab3.pdf Restricted to Registered users only Download (257kB) | Request a copy
	Text (BAB 4) 7_bab4.pdf Restricted to Registered users only Download (348kB) | Request a copy
	Text (BAB 5) 8_bab5.pdf Restricted to Registered users only Download (89kB) | Request a copy
	Text (DAFTAR PUSTAKA) 9_daftarpustaka.pdf Restricted to Registered users only Download (105kB) | Request a copy

Abstract

INDONESIA : Rata-rata waktu tunggu mendapatkan pekerjaan pertama bagi Sarjana adalah nol sampai 9 bulan. Berdasarkan data dari Badan Pusat Statistik, pada saat ini tingkat pendidikan tinggi yang telah ditempuh seseorang tidak menjamin mudahnya mendapatkan pekerjaan. Oleh karena itu, perlu untuk mencari tahu apasaja faktor yang dapat memengaruhi lulusan sarjana dalam mendapatkan pekerjaan pertamanyan dengan menggunakan analisis survival. Analisis survival merupakan analisis ketahanan hidup dengan menganalisis data yang berupa waktu antar kejadian. Terdapat banyak model yang bisa digunakan, diantaranya adalah model Regresi Cox Proportional Hazard dan model Accelerated Failure Time (AFT). Model Accelerated Failure Time termasuk kedalam model statistika parametrik yang mana membutuhkan distribusi data tertentu. Distribusi data yang dapat digunakan, diantaranya ialah distribusi Weibull, Eksponensial, Log-Logistik, Log-Normal. Keempat distribusi tersebut digunakan dalam penelitian kali ini untuk pengolahan data. Sedangkan model CPH tidak membutuhkan distribusi data tertentu karena termasuk kedalam model statistika semi parametrik. Untuk mendapatkan model terbaik dari Model Regresi Cox Proportional Hazard dan model AFT dapat dilakukan dengan melihat nilai AIC atau BIC terkecil. Pada kasus waktu tunggu sarjana, dengan objek penelitian yaitu alumni angkatan 2014, 2015, 2016, dan 2017 di Fakultas Sains dan Teknologi UIN Sunan Gunung Djati Bandung. Model AFT yang berdistribusi log-Normal lebih baik dibandingkan model lainnya dan variabel yang berpengaruh secara signifikan ialah jenis kelamin. ENGLISH : The average waiting time to get the first job for a Bachelor is zero to 9 months. Based on data from the Central Statistics Agency, at this time the level of higher education that a person has taken does not guarantee the ease of getting a job. Therefore, it is necessary to find out what are the factors that can influence undergraduate graduates in getting their first job using survival analysis. Survival analysis is a survival analysis by analyzing data in the form of time between events. There are many models that can be used, including the Cox Proportional Hazard Regression model and the Accelerated Failure Time (AFT) model. The Accelerated Failure Time model is included in the parametric statistical model which requires a certain data distribution. Data distributions that can be used, including Weibull, Exponential, Log-Logistics, Log-Normal distributions. The four distributions were used in this study for data processing. Meanwhile, the CPH model does not require a certain data distribution because it is included in a semi-parametric statistical model. To get the best model of the Cox Proportional Hazard Regression Model and the AFT model can be done by looking at the smallest AIC or BIC values. In the case of undergraduate waiting time, with the object of research, namely alumni of the class of 2014, 2015, 2016, and 2017 at the Faculty of Science and Technology UIN Sunan Gunung Djati Bandung. The Normal-log-distributed AFT model is better than other models and the variable that has a significant effect is gender.

Actions (login required)

View Item