Penerapan Model Poisson Generalized Autoregressive Moving Average (GARMA) Untuk Peramalan Jumlah Penderita Demam Berdarah Dengue (DBD)

Aisyah, Aurora Hanun (2022) Penerapan Model Poisson Generalized Autoregressive Moving Average (GARMA) Untuk Peramalan Jumlah Penderita Demam Berdarah Dengue (DBD). Other thesis, Institut Teknologi Sepuluh Nopember.

Text 06111840000008-Undergraduate_Thesis.pdf Restricted to Repository staff only Download (1MB) | Request a copy

Abstract

Demam Berdarah Dengue (DBD) merupakan penyakit yang disebabkan oleh virus dengue. Penyakit ini ditularkan oleh nyamuk Aedes sp. yang telah terinfeksi virus dengue. Jumlah penderita DBD di Indonesia termasuk di Kabupaten Lamongan selalu terjadi fluktuasi. Oleh karena itu dilakukan peramalan jumlah penderita DBD di Kabupaten Lamongan pada tahun 2015 sampai tahun 2021. Data jumlah penderita DBD di Kabupaten Lamongan merupakan data count. Salah satu metode peramalan yang banyak digunakan adalah ARIMA tetapi pada kasus data count, ARIMA tidak selalu tepat digunakan. Sehingga digunakan model Generalized Autoregressive Moving Average (GARMA) dengan data jumlah penderita DBD di Kabupaten Lamongan yang mengikuti distribusi Poisson maka disebut juga Poisson GARMA (p, q). Data yang digunakan pada penelitian ini diketahui tidak bersifat musiman sehingga model Poisson GARMA yang digunakan pada penelitian ini tidak melibatkan efek musiman. Estimasi parameter yang dilakukan pada penelitian ini menggunakan metode Maximum Likelihood Estimation (MLE) dan dilanjutkan dengan dengan algoritma Iteratively Reweighted Least Squares (IRLS). Hasil yang didapat pada penelitian ini adalah model terbaik untuk data tersebut, yaitu model Poisson GARMA (1, 1). Pemilihan model terbaik berdasarkan nilai Akaike’s Information Criterion (AIC) terkecil yaitu sebesar 216, 3983. Model Poisson GARMA (1, 1) diaplikasikan pada data jumlah penderita DBD di K
===================================================================================================================================
Dengue Hemorrhagic Fever (DHF) is a disease caused by the dengue virus. This disease transmitted by the Aedes sp. which is infected with the dengue virus. The number of DHF patients in Indonesia, including in the Lamongan Regency, always fluctuates. Therefore, it is necessary to forecast the number of DHF patients in Lamongan Regency in 2015 to 2021. The number of DHF patients in Lamongan Regency data is the count data. One of the widely used forecasting methods is ARIMA but in the case of count data, it is not always appropriate to use it. So that the Generalized Autoregressive Moving Average (GARMA) model is used with the number of DHF patients in Lamongan Regency data that follows the Poisson distribution, it is also called Poisson GARMA (p, q). The data used in this research is known that the data are not seasonal so the Poisson GARMA model used in this research does not involve seasonal effects. In this research, using the Maximum Likelihood Estimation (MLE) method followed by Iteratively Reweighted Least Squares (IRLS) algorithm for the parameter estimation. The results of this research are the best model for the data is the Poisson GARMA model (1.1). The best model selection based on the smallest value of Akaike’s Information Criterion (AIC) that is 216.3983. The Poisson GARMA (1.1) model was applied to the data on the number of DHF patients in Lamongan Regency to get a forecast number for the next one-year period.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Data Count, DBD, IRLS, Poisson GARMA.
Subjects:	Q Science > QA Mathematics > QA274.7 Markov processes--Mathematical models.
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Mathematics > 44201-(S1) Undergraduate Thesis
Depositing User:	Mr. Marsudiyana -
Date Deposited:	22 Oct 2025 06:49
Last Modified:	22 Oct 2025 06:49
URI:	http://repository.its.ac.id/id/eprint/128659

Actions (login required)

View Item