Generalized Space Time Autoregressive (GSTAR) Model for Air Temperature Forecasting in the South Sumatera, Riau, and Jambi Provinces
Abstract
Over the past few years, there has been a significant increase in air temperatures in regions such as South Sumatera, Riau, and Jambi, posing threats of drought, water resource crises, and erratic weather patterns. In response, developing air temperature forecasting techniques becomes imperative for effective climate change management. This study proposes implementing the Generalized Space Time Autoregressive (GSTAR) model as a practical approach for forecasting air temperatures in these regions using two weighting methods, i.e., inverse distance and normalized cross-correlation weighting. The GSTAR model, an extension of the Space Time Autoregressive (STAR) model, offers enhanced complexity by incorporating specific time and location factors, thereby increasing forecasting flexibility. The result reveals that GSTAR(1,1) with normalized cross-correlation weighting is the most optimal model, with a Root Mean Square Error (RMSE) value of 3.135, indicating high forecasting accuracy. The selection of this model is grounded in the geographical proximity and similarity of environmental characteristics of the three regions. This research contributes novel insights into the underlying mechanisms of air temperature dynamics in neighboring areas, providing a robust foundation for formulating effective policy and mitigation strategies in addressing climate change challenges.
Keywords: air temperatures, normalized cross-correlation weighting, GSTAR(1,1), inverse distance weighting.
Abstrak
Dalam beberapa tahun terakhir, suhu udara mengalami peningkatan signifikan di wilayah-wilayah seperti Sumatera Selatan, Riau, dan Jambi, yang mengancam kekeringan, krisis sumber daya air, dan perubahan pola cuaca yang tidak terduga. Menghadapi situasi tersebut, pengembangan teknik peramalan suhu udara diperlukan untuk mengantisipasi dan mengelola dampak ekstrem dari perubahan iklim. Studi ini mengusulkan implementasi model Generalized Space Time Autoregressive (GSTAR) sebagai pendekatan praktis untuk meramalkan suhu udara di wilayah-wilayah tersebut menggunakan dua metode pembobotan yaitu pembobotan invers jarak dan normali korelasi silang. Model GSTAR, sebagai perluasan dari model Space Time Autoregressive (STAR), menawarkan kompleksitas yang lebih baik dengan menggabungkan faktor-faktor waktu dan lokasi tertentu, sehingga meningkatkan fleksibilitas dalam ramalan. Hasil analisis menunjukkan bahwa GSTAR(1,1) dengan pemberian bobot normalisasi korelasi silang merupakan model yang paling optimal, dengan nilai Root Mean Square Error (RMSE) sebesar 3.135, menandakan tingkat akurasi yang tinggi. Pemilihan model ini didasarkan pada kedekatan geografis dan kesamaan karakteristik lingkungan dari ketiga wilayah tersebut. Penelitian ini memberikan wawasan baru dalam mekanisme dinamika suhu udara di wilayah-wilayah yang berdekatan, serta memberikan dasar yang kuat bagi perumusan kebijakan dan strategi mitigasi yang efektif dalam menghadapi tantangan perubahan iklim.
Kata Kunci: bobot invers jarak, bobot normalisasi korelasi silang, GSTAR(1,1), suhu udara.
2020MSC: 62P30
Keywords
References
G. I. Lippsmeier, Tropenbau Building in the Tropics. German: Callway, 1980.
C. Chou, J. D. Neelin, C. A. Chen, and J. Y. Tu, “Evaluating the ‘rich-get-richer’ mechanism in tropical precipitation change under global warming,” J. Clim., vol. 22, no. 8, pp. 1982–2005, 2009, doi: 10.1175/2008JCLI2471.1.
A. Machmudin and B. S. S. Ulama, “Peramalan Temperatur Udara di Kota Surabaya dengan Menggunakan ARIMA dan Artificial Neural Network,” J. Sains Dan Seni ITS, vol. 1, no. 1, pp. 118–123, 2012.
A. F. Hadi, I. Yudistira, D. Anggraeni, and M. Hasan, “The Geographical Clustering of the Rainfall Stations on Seasonal GSTAR Modeling for Rainfall Forecasting,” J. Phys. Conf. Ser., vol. 1028, no. 1, 2018, doi: 10.1088/1742-6596/1028/1/012238.
G. Kirchgassner and J. Wolters, Introduction to Modern Time Series Analysis. New York: Springer, 2007.
M. Usman et al., “Analysis of Some Energy and Economics Variables by Using VECMX Model in Indonesia,” Int. J. Energy Econ. Policy, vol. 12, no. 2, pp. 91–102, 2022, doi: 10.32479/ijeep.11897.
P. E. Pfeifer and S. J. Deutsch, “A Three-Stage Iterative Procedure for Space-Time Modeling,” Technometrics, vol. 22, no. 1, p. 35, 1980, doi: 10.2307/1268381.
B. N. Ruchjana, “PEMODELAN KURVA PRODUKSI MINYAK BUM1 MENGGUNAKAN MODEL GENERALISASI S-TAR,” Forum Stat. dan komputasi, pp. 1–6, 2002.
D. R. S. Saputro, I. M. Putri, Sutanto, N. H. Noor, and P. Widyaningsih, “Generalized space time autoregressive (gstar)-artificial neural network (ann) model with multilayer feedforward networks architecture,” IOP Conf. Ser. Earth Environ. Sci., vol. 243, no. 1, 2019, doi: 10.1088/1755-1315/243/1/012039.
A. P. Muzdhalifah, T. Tarno, and P. Kartikasari, “Penerapan Model Generalized Space Time Autoregressive (Gstar) Untuk Meramalkan Penerbangan Domestik Pada Tiga Bandar Udara Di Pulau Jawa,” J. Gaussian, vol. 11, no. 3, pp. 332–343, 2023, doi: 10.14710/j.gauss.11.3.332-343.
N. Ilmi, A. Aswi, and M. K. Aidid, “Generalized Space Time Autoregressive Integrated Moving Average (GSTARIMA) dalam Peramalan Data Curah Hujan di Kota Makassar,” Inferensi, vol. 6, no. 1, p. 25, 2023, doi: 10.12962/j27213862.v6i1.14347.
S. Borovkova, H. P. Lopuhaä, and B. N. Ruchjana, “Consistency and asymptotic normality of least squares estimators in generalized STAR models,” Stat. Neerl., vol. 62, no. 4, pp. 482–508, 2008, doi: 10.1111/j.1467-9574.2008.00391.x.
B. N. Ruchjana, S. A. Borovkova, and H. P. Lopuhaa, “Least squares estimation of Generalized Space Time AutoRegressive (GSTAR) model and its properties,” AIP Conf. Proc., vol. 1450, no. May, pp. 61–64, 2012, doi: 10.1063/1.4724118.
D. N. Gujarati and D. C. Porter, The McGraw-Hill Series Economics. 2009.
N. M. Huda and N. Imro’ah, “Determination of the best weight matrix for the Generalized Space Time Autoregressive (GSTAR) model in the Covid-19 case on Java Island, Indonesia,” Spat. Stat., vol. 54, p. 100734, 2023, doi: 10.1016/j.spasta.2023.100734.
Suhartono and Subanar, “The Optimal Determination of Space Weight in GSTAR Model by Using Cross-correlation Inference,” J. Quant. Method, J. Devoted to Math. Stat. Apl. Var. F., vol. 2, no. 2, pp. 45–53, 2006.
I. Adam, D. Kusnandar, and H. Perdana, “Penerapan Model GSTAR(1,1) Untuk Data Curah Hujan,” Bul. Ilm. Math. Stat. dan Ter., vol. 6, no. 3, pp. 159–166, 2017.
S. S. Handajani, H. Pratiwi, Y. Susanti, S. Subanti, Respatiwulan, and Hartatik, “Rainfall model on area of rice production in Sragen, Karanganyar and Klaten by using Generalized Space Time Autoregressive (GSTAR),” J. Phys. Conf. Ser., vol. 855, no. 1, 2017, doi: 10.1088/1742-6596/855/1/012015.
U. S. Pasaribu, U. Mukhaiyar, N. M. Huda, K. N. Sari, and S. W. Indratno, “Modelling COVID-19 growth cases of provinces in java Island by modified spatial weight matrix GSTAR through railroad passenger’s mobility,” Heliyon, vol. 7, no. 2, p. e06025, 2021, doi: 10.1016/j.heliyon.2021.e06025.
F. N. Aryani, S. S. Handajani, and E. Zukhronah, “Penerapan Model Generalized Space Time Autoregressive Pada Data Nilai Tukar Petani Di Tiga Provinsi Pulau Sumatera,” J. Ilmu Dasar, pp. 97–104, 2020.
C. M. Hurvich and C. L. Tsai, “Regression and time series model selection in small samples,” Biometrika, vol. 76, no. 2, pp. 297–307, 1989, doi: 10.1093/biomet/76.2.297.
A. Fadlurohman, T. Wahyu Utami, R. Wasono, K. Kunci, J. Tengah, and S. Biaya Hidup, “Generalized Space Time Autoregressive Modeling With Variable Exogenous (Gstar-X) (Case Study: Inflation In Six Cities Of Central Java),” Pros. Semin. Nas. Unimus, vol. 3, pp. 26–36, 2020.
W. W. S. Wei, Multivariate Time Series Analysis and Application. New York: John Wiley and Sons, Inc, 2019.
DOI: 10.15408/inprime.v6i1.36049
Refbacks
- There are currently no refbacks.
.
