Penerapan Model Generalized Space Time Autoregressive (GSTAR) pada Data Inflasi Beberapa Kota
Abstract
Model yang umum digunakan untuk data space time adalah model Vector autoregressive (VAR), Space Time Autoregressive (STAR), dan Generalized Space Time Autoregressive (GSTAR). Untuk lokasi yang memiliki karakteristik yang berbeda (heterogen), model GSTAR lebih baik digunakan dibandingkan model STAR. Tujuan dari penelitian ini adalah menerapkan model GSTAR pada data time series dari tiga lokasi berbeda. Data yang digunakan pada penelitian ini adalah data inflasi Palembang, Bandar Lampung, dan DKI Jakarta bulan Januari 2012 hingga Juni 2019. Bobot Lokasi yang digunakan adalah bobot lokasi invers jarak dan bobot lokasi normalisasi korelasi silang. Pada penelitian ini pendugaan parameter dilakukan dengan metode Generalized Least Square (GLS). Dari hasil analisis diperoleh model yang terbaik adalah model GSTAR(11) dengan bobot lokasi invers jarak karena memiliki rata-rata RMSE terkecil yaitu 0.467767.
The models commonly used for space time data are the Vector autoregressive (VAR), Space Time Autoregressive (STAR), and Generalized Space Time Autoregressive (GSTAR) models. For locations that have different (heterogeneous) characteristics, the GSTAR model is better to use than the STAR model. The aim of this research is to apply the GSTAR model to time series data from three different locations. The data used in this research is inflation data from Palembang, Bandar Lampung, and DKI Jakarta from January 2012 to June 2019. The location weights used are distance inverse location weights and cross-correlation normalized location weights. In this research, parameter estimation was carried out using the Generalized Least Square (GLS) method. From the analysis results, it was found that the best model was the GSTAR(11) model with inverse distance location weights because it had the smallest average RMSE, namely 0.467767.
References
Al Amri, M. Z. (2020). Perbandingan Model STAR dan GSTAR untuk Peramalan Indeks Harga Konsumen di Kota Padang, Pekanbaru, Jambi, dan Palembang. Media Edukasi Data Ilmiah Dan Analisis (MEDIAN), 3(01), 29–38.
Arum, P. R., Fathoni Amri, I., & Amri, S. (2024). GLS estimation in python to forecast gross regional domestic product using generalized space–time autoregressive seemingly unrelated regression model. Frontiers in Applied Mathematics and Statistics, 10, 1365723. https://doi.org/10.3389/fams.2024.1365723
Bi, J., Wang, J., Cao, H., Yao, G., Wang, Y., Li, Z., Sun, M., Yang, H., Zhen, J., & Zheng, G. (2024). Inverse distance weight-assisted particle swarm optimized indoor localization. Applied Soft Computing, 164, 112032. https://doi.org/10.1016/j.asoc.2024.112032
Borovkova, S. A., Lopuhaä, H. P., & Nurani, B. (2002). Generalized STAR model with experimental weights. Proceedings of the 17th International Workshop on Statistical Modelling, 139–147.
Fransiska, H., Sunandi, E., & Agustina, D. (2020). Peramalan Curah Hujan Provinsi Bengkulu dengan Generalized Space-Time Autoregressive. MUST: Journal of Mathematics Education, Science and Technology, 5(2), 130–142. https://doi.org/10.30651/must.v5i2.5326
Hestuningtias, F., & Kurniawan, M. H. S. (2023). The Implementation of the Generalized Space-Time Autoregressive (GSTAR) Model for Inflation Prediction. Enthusiastic: International Journal of Applied Statistics and Data Science, 176–188. https://doi.org/10.20885/enthusiastic.vol3.iss2.art5
Imro’ah, N. (2023). Determination of the best weight matrix for the Generalized Space Time Autoregressive (GSTAR) model in the Covid-19 case on Java Island, Indonesia. Spatial Statistics, 54, 100734. https://doi.org/10.1016/j.spasta.2023.100734
Iriany, A., Aini, N. N., & Sulistyono, A. D. (2021). Spatio Temporal Modelling for Government Policy the COVID-19 Pandemic in East Java. CAUCHY: Jurnal Matematika Murni Dan Aplikasi, 6(4), 218–226. https://doi.org/10.18860/ca.v6i4.10639
Mukhaiyar, U., & Ramadhani, S. (2022). The Generalized STAR Modeling with Heteroscedastic Effects. CAUCHY: Jurnal Matematika Murni Dan Aplikasi, 7(2), 158–172. https://doi.org/10.18860/ca.v7i2.13097
Ord, J. K. (2010). Spatial autocorrelation: A statistician’s reflections. Perspectives on Spatial Data Analysis, 165–180. https://doi.org/10.1007/978-3-642-01976-0_12
Rompon, M., Soblia, H. T., Monika, P., Abdullah, A. S., & Ruchjana, B. N. (2023). Identifikasi Autokorelasi Spasial Warisan Budaya Tak Benda di Indonesia Menggunakan Indeks Moran. Statistika, 23(2), 156–163. https://doi.org/10.29313/statistika.v23i2.2675
Ruiz, A. P., Flynn, M., Large, J., Middlehurst, M., & Bagnall, A. (2021). The great multivariate time series classification bake off: a review and experimental evaluation of recent algorithmic advances. Data Mining and Knowledge Discovery, 35(2), 401–449. https://doi.org/10.1007/s10618-020-00727-3
Woodward, W. A., Gray, H. L., & Elliott, A. (2020). Nonstationary Time Series Models. In Applied Time Series Analysis (pp. 203–220). https://doi.org/10.1201/b11459-9
Yilmaz, V. (2023). Decomposition of effects for the structural model consisting of two mediating latent variables: an example of entrepreneurial intention. Journal of Modelling in Management, 18(3), 973–992. https://doi.org/10.1108/JM2-01-2022-0008
Zhao, L., Li, Z., & Qu, L. (2022). Forecasting of Beijing PM2. 5 with a hybrid ARIMA model based on integrated AIC and improved GS fixed-order methods and seasonal decomposition. Heliyon, 8(12). https://doi.org/10.1016/j.heliyon.2022.e12239
