DOI: https://doi.org/10.14710/j.gauss.14.2.247-256

PEMODELAN REGRESI GAMMA MENGGUNAKAN METODE OPTIMASI BROYDEN-FLETCHER-GOLDFARB-SHANNO (BFGS) (Studi Kasus : Pencemaran Sungai di Kota Semarang)

Efifah Nur Safitri  -  Departemen Statistika, Fakultas Sains dan Matematika, Universitas Diponegoro, Indonesia
*Arief Rachman Hakim  -  Departemen Statistika, Fakultas Sains dan Matematika, Universitas Diponegoro, Indonesia
Suparti Suparti  -  Departemen Statistika, Fakultas Sains dan Matematika, Universitas Diponegoro, Indonesia
Open Access Copyright 2025 Jurnal Gaussian under http://creativecommons.org/licenses/by-nc-sa/4.0.

Citation Format:
Abstract
Semarang City is one of the major industrial areas in Central Java and is located not far from residential areas. These large industries utilize a variety of chemicals to meet their needs, producing various wastes that are the main cause of high concentrations of Chemical Oxygen Demand (COD) in waters. Measurements on the COD value obtained are continuous data with Gamma distribution. In this study, Gamma regression is used to model the relationship between one or more predictor variables and a positive continuous response variable following a Gamma distribution. Parameter estimation in the Gamma regression model uses the Maximum Likelihood Estimation (MLE) method because it does not produce an analytical solution, an optimization method will be carried out with the Broyden-Fletcher-Goldfarb-Shanno (BFGS) algorithm. This study aims to determine the Gamma regression model in the case of river pollution in Semarang City, determine the form of parameter estimation in the Gamma regression model and get what factors affect river pollution in Semarang City. Based on the  value of 0.4431498 where the ability of the predictor variables to explain the response variable is 44.32%, the remaining 55.68% of the response variable is explained by other factors not contained in the model.
Fulltext View|Download
Keywords: River Pollution; Gamma Regression; COD; MLE; BFGS.

Article Metrics:

Article Info
Section: Articles
Language : EN
Recent articles
PERAMALAN HARGA BERAS DI INDONESIA MENGGUNAKAN METODE HOLT-WINTERS ADDITIVE EXPONENTIAL SMOOTHING DENGAN OPTIMASI GOLDEN SECTION Penerapan Metode Fuzzy Time Series Markov Chain Untuk Meramalkan Nilai Transaksi Belanja Menggunakan Uang Elektronik di Indonesia PERBANDINGAN PERFORMA MODEL ARIMA-GARCH DAN LSTM DALAM MERAMALKAN JUMLAH KUNJUNGAN WISATAWAN DANAU KASTOBA More recent articles
  1. Bossio, M. C., & Cuervo, E. C. 2015. Gamma regression models with the Gammareg R package. Comunicaciones en estadistica, 8(2), 211-223
  2. Dobson, A. J., & Barnett, A. G. 2018. An introduction to generalized linear models. CRC press
  3. Putri, E. D., & Prastyo, D. 2017. Parameter estimation and hypothesis testing on geographically weighted gamma regression. In Journal of Physics Conference Series (Vol. 893, No. 1, p. 012025)
  4. Faizah, K. 2018. Regresi Gamma dan Regresi Lognormal untuk Memodelkan Biomassa Klorofil-a di Sungai Danube Eropa. Tesis. Institut Teknologi Sepuluh November Surabaya
  5. Ghozali, I. 2018. Aplikasi Analisis Multivariate dengan Program IBM SPSS 25. Badan Penerbit Universitas Diponegoro: Semarang
  6. Gujarati, D. N. 2004. Basic Econometrics (4 ed.). New York: The McGraw-Hill
  7. Lumaela, A. K., Otok, B. W., & Sutikno, S. 2013. Pemodelan chemical oxygen demand (cod) sungai di Surabaya dengan metode mixed geographically weighted regression. Jurnal Sains dan Seni ITS, 2(1), D100-D105
  8. Luthfiana, U. 2021. Model Regresi Gamma Pada Indikator Pencemaran Sungai di Kota Semarang. Skripsi. Universitas Islam Negeri Walisongo Semarang
  9. Misbahussurur, A. 2009. Estimasi Parameter Distribusi Gamma dengan Metode Maksimum Likelihood. Skripsi. Universitas Islam Negeri Maulana Malik Ibrahim
  10. Nasution, A.S. 2017. Estimasi Parameter dan Pengujian Hipotesis pada Model Regresi Gamma (Studi Kasus Pemodelan Pencemaran Air Sungai di Surabaya). Tesis. Institute Teknologi Sepuluh Nopember Surabaya
  11. Saputro, D. R., & Purnami, W. 2016. Algoritme Pendugaan Parameter Model Regresi Logistic Biler (RLB) dengan Maksimum Likelihood dan Broyden-Fletcher-Goldfarb-Shanno (BFGS). Seminar Nasional Matematika Dan Pendidikan Matematika

Last update:

No citation recorded.

Last update:

No citation recorded.