Evaluation of the Arima-Kalman model in predicting rainfall in Medan City in 2023 using observation data from 2013 – 2022

Alva Josia Lumbantoruan, Yahya Darmawan, Munawar Munawar, Nardi Nardi, Fendy Arifianto, Ervan Ferdiansyah

Abstract


This paper aims to evaluate the ARIMA-Kalman model in predicting rainfall in Medan City for the year 2023. The data used are historical observation data of rainfall from 2013 to 2022 that have been tested for stationary and homogeneity, which proved not to require additional correction. The analysis results show that the ARIMA-Kalman model can capture the general pattern of rainfall well, and shows superiority in producing predictions that are closer to the actual data, with a mean absolute error (MAE) value of 54.11, which is lower than the MAE of the ARIMA model which reaches 55.66. Although the ARIMA model has a smaller root mean square error (RMSE) (66.67 compared to 69.75 for ARIMA-Kalman), the ARIMA-Kalman model shows better consistency, especially in capturing significant fluctuations, such as the peak rainfall that occurred in July 2023. Therefore, ARIMA-Kalman is proven to be more accurate and reliable for predicting rainfall in Medan city, making it a better choice to support water resources planning and management.

Keywords


ARIMA; Kalman filter; model; prediction; rainfall

Full Text:

PDF

References


1. Siregar, N. A. (2022). Peramalan curah hujan di Kota Medan menggunakan metode support vector regression. J. Informatics Data Sci., 1(1), 7–9.

2. Anggraini, N., Pangaribuan, B., Siregar, A. P., Sintampalam, G., Muhammad, A., Damanik, M. R. S., & Rahmadi, M. T. (2021). Analisis pemetaan daerah rawan banjir di Kota Medan Tahun 2020. Jurnal Samudra Geografi, 4(2), 27–33.

3. Selamat, I. K. & Setyawan, Y. (2023). Prediksi curah hujan perbulan di Kota Yogyakarta Periode 2015 – 2019 menggunakan metode autoregressive integrated moving average (ARIMA) dan Kalman filter (Studi kasus : Data curah hujan Tahun 2015 – 2019). J. Stat. Ind. dan Komputasi, 8(1), 15–31.

4. Aprilia, M. & Desviona, N. (2021). The implementation of a filter Kalman method forecasting rainfall obtained through model ARIMA in Kota Jambi. Nucleus, 2(2), 69

5. Zulfi, M., Hasan, M., & Purnomo, K. D. (2018). The development rainfall forecasting using Kalman filter. Journal of Physics: Conference Series, 1008(1), 012006.

6. Ananda, E. Y. P. & Wahyuni, M. S. (2021). Rainfall forecasting model using ARIMA and Kalman filter in Makassar, Indonesia. Journal of Physics: Conference Series, 2123(1), 012044.

7. Rusdi, R. (2011). Uji Akar-Akar Unit dalam Model Runtun Waktu Autoregresif. Statistika, 11(2), 67–78.

8. Ruhiat, D., Andiani, D., & Kamilah, W. N. (2020). Forecasting data runtun waktu musiman menggunakan metode singular spectrum analysis (SSA). Teorema: Teori dan Riset Matematika, 5(1), 47–60.

9. Samsiah, D. N. (2008). Analisis data runtun waktu menggunakan model ARIMA (p, d, q). Skripsi UIN Sunan Kalijaga, Yogyakarta.

10. Laamena, N. S. (2019). Prakiraan harga rumah di Kota Semarang dengan model deret waktu. Jurnal Satya Informatika, 4(01), 41–52.

11. Rianto, M. & Yunis, R. (2021). Analisis runtun waktu untuk memprediksi jumlah mahasiswa baru dengan model random forest. Paradigma, 23(1), 70–74.




DOI: http://dx.doi.org/10.31258/jkfi.22.1.15-22

Refbacks

  • There are currently no refbacks.


Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.

Indexing by:

  

 

Image