Sains Malaysiana 53(11)(2024): 3817-3829

http://doi.org/10.17576/jsm-2024-5311-23

 

Modelling Malaysia Air Quality Data using Bayesian Structural Time Series Models

(Memodelkan Data Kualiti Udara Malaysia menggunakan Model Siri Masa Berstruktur Bayesian)

 

AESHAH MOHAMMED^1,2, MOHD AFTAR ABU BAKAR^1,*, MAHAYAUDIN M. MANSOR³ & NORATIQAH MOHD ARIFF¹

 

¹Department of Mathematical Sciences, Faculty of Science and Technology, Universiti Kebangsaan Malaysia, 43600 UKM Bangi, Selangor, Malaysia
²Faculty of Science, University of Benghazi, AL Marje, Libya
 ³School of Mathematical Sciences, College of Computing, Informatics and Mathematics, Universiti Teknologi MARA, 40450 Shah Alam, Selangor, Malaysia

 

Received: 9 February 2024/Accepted: 27 September 2024

 

Abstract

Air pollution poses a significant threat to human health and the environment, especially in developing nations facing rapid industrialization, urbanization, and increased vehicle emissions. As cities and factories continue to grow, the air quality problem worsens, making it crucial to enhance the monitoring, testing, and forecasting of air quality. In this context, this study focuses on building air quality models using Bayesian Structural Time Series (BSTS) models to predict air quality levels in Malaysia. The BSTS model integrates three main techniques: The structural model, which employs the Kalman filter approach to model trend and seasonality components; spike and slab regression for variable selection; and Bayesian model averaging to estimate the best-performing prediction model while accounting for uncertainty. The study utilized air quality time-series data spanning two years, from June 2017 to July 2019, obtained from the Malaysian Department of Environment (DOE). The primary objective of this study was to forecast air quality and assess the effectiveness of the Bayesian structural time series analysis on air quality time-series data. The results indicated that the BSTS technique is capable of modeling air quality time-series data with high accuracy, effectively capturing seasonal and trend components. The seasonal component showed a repetition of weekly concentration patterns, while the local linear trend component showed a steady decline in PM₁₀ and PM_2.5 concentration levels in most stations. Regression analysis demonstrated that humidity and ambient temperature significantly affected air quality in most locations in Malaysia.

 

Keywords: Air quality; Bayesian Structural Time Series; Monte Carlo Markov Chain (MCMC); spike and slab regression

 

Abstrak

Pencemaran udara menimbulkan ancaman besar kepada kesihatan manusia dan alam sekitar, terutamanya di negara membangun yang menghadapi perindustrian pesat, pembandaran dan peningkatan pelepasan kenderaan. Perkembangan bandar dan pertambahan kilang mengakibatkan masalah kualiti udara bertambah buruk, menjadikan pentingnya pemantauan, ujian dan ramalan kualiti udara. Dalam konteks ini, kajian ini tertumpu kepada pembinaan model kualiti udara menggunakan model Siri Masa Berstruktur Bayesian (BSTS) untuk meramalkan tahap kualiti udara di Malaysia. Model BSTS menyepadukan tiga teknik utama: Model struktur yang menggunakan pendekatan penapis Kalman untuk memodelkan komponen trend dan bermusim; regresi spike dan papak untuk pemilihan berubah; dan model Bayesian secara purata untuk menganggarkan model ramalan berprestasi terbaik sambil mengambil kira ketidakpastian. Kajian itu menggunakan data siri masa kualiti udara yang menjangkau dua tahun dari Jun 2017 hingga Julai 2019 yang diperoleh daripada Jabatan Alam Sekitar Malaysia (JAS). Objektif utama kajian ini adalah untuk meramal kualiti udara dan menilai keberkesanan analisis BSTS terhadap data siri masa kualiti udara. Keputusan menunjukkan bahawa teknik BSTS mampu memodelkan data siri masa kualiti udara dengan ketepatan yang tinggi, menangkap komponen bermusim dan trend dengan berkesan. Komponen bermusim menunjukkan pengulangan corak kepekatan mingguan, manakala komponen aliran linear tempatan menunjukkan penurunan yang stabil dalam tahap kepekatan PM₁₀ dan PM_2.5 di kebanyakan stesen. Analisis regresi menunjukkan bahawa kelembapan dan suhu ambien menjejaskan kualiti udara dengan ketara di kebanyakan lokasi di Malaysia.

 

Kata kunci: Kualiti udara; Rantaian Markov Monte Carlo (MCMC); regresi pepaku dan papak; Siri Masa Berstruktur Bayesian

 

REFERENCES

Almarashi, A.M. & Khan, K. 2020. Bayesian structural time series. Nanoscience and      Nanotechnology Letters 12(1): 54-61. 

Ariff, N.M., Bakar, M.A.A. & Lim, H.Y. 2023. Prediction of PM₁₀ concentration in Malaysia using k-means clustering and LSTM hybrid model. Atmosphere 14(5): 853.

Bakar, M.A.A., Ariff, N.M., Bakar, S.A., Chi, G.P. & Rajendran, R. 2022. Peramalan kualiti udara menggunakan kaedah pembelajaran mendalam rangkaian perlingkaran temporal (TCN). Sains Malaysiana 51(8): 2645-2654.

Bakar, M.A.A., Mohd Ariff, N.M., Mohd Nadzir, M.S., Wen, O.L. & Suris, F.N.A. 2022. Prediction of multivariate air quality time series data using long short-term memory network. Malaysian Journal of Fundamental and Applied Sciences 18(1): 52-59.

Brodersen, K.H., Gallusser, F., Koehler, J., Remy, N. & Scott, S.L. 2015. Inferring causal impact using bayesian structural time-series models. The Annals of Applied Statistics 9(1): 247-274.

Durbin, J. & Koopman, S.J. 2002. A simple and efficient simulation smoother for state space time series analysis. Biometrika 89(3): 603-615.

George, E. & McCulloch, R. 1997. Approaches for Bayesian variable selection. Statistica Sinica 7(2): 339-373.

Jun, S. 2019. Bayesian structural time series and regression modeling for sustainable technology management. Sustainability (Switzerland) 11(18): 4945.

Kalman, R.E. 1960. A new approach to linear filtering and prediction problems. Journal of Fluids Engineering, Transactions of the ASME 82(1): 35-45.

Madigan, D. & Raftery, A.E. 1994. Model selection and accounting for model uncertainty in graphical models using occam’s window. Journal of the American Statistical Association 89(428): 1535-1546.

Mokilane, P., Debba, P., Yadavalli, V. & Sigauke, C. 2019. Bayesian structural time-series approach to a long-term electricity demand forecasting. Applied Mathematics and Information Sciences 13: 189-199.

Mun, C.K., Abd Rahman, N.H. & Che Ilias, I.S. 2022. Performance of Levenberg-Marquardt neural network algorithm in air quality forecasting. Sains Malaysiana 51(8): 2645-2654.

Nasr Ahmed AL-Dhurafi, Nurulkamal Masseran & Zamira Hasanah Zamzuri. 2018. Compositional time series analysis for air pollution index data. Stochastic Environmental Research and Risk Assessment 32(10): 2903-2911. https://doi.org/10.1007/s00477-018-1542-0

Nurulkamal Masseran & Muhammad Aslam Mohd Safari. 2020. Modeling the transition behaviors of PM₁₀ pollution index. Environmental Monitoring and Assessment 192: 441. https://api.semanticscholar.org/CorpusID:219729578

Nurul Nnadiah Zakaria, Mahmod Othman, Rajalingam Sokkalingam, Hanita Daud, Lazim Abdullah & Evizal Abdul Kadir. 2019. Markov chain model development for forecasting air pollution index of Miri, Sarawak. Sustainability (Switzerland) 11(19): 5190. https://doi.org/10.3390/su11195190

Scott, S.L. & Varian, H.R. 2014. Predicting the present with Bayesian structural time series. International Journal of Mathematical Modelling and Numerical Optimisation 5(1-2): 4-23.

Volinsky, C.T., Raftery, A.E., Madigan, D. & Hoeting, J.A. 1999. David Draper and E.I. George, and a rejoinder by the authors. Statistical Science 14(4): 382-417.

Wen, Z., Ma, X., Xu, W., Si, R., Liu, L., Ma, M., Zhao, Y., Tang, A., Zhang, Y., Wang, K., Zhang, Y., Shen, J., Zhang, L., Zhao, Y., Zhang, F., Goulding, K. & Liu, X. 2024. Combined short-term and long-term emission controls improve air quality sustainably in China. Nature Communications 15(1): 5169.

Zellner, A. 1986. On assessing prior distributions and Bayesian regression analysis with g-prior distributions. In Bayesian Inference and Decision Techniques: Essays in Honor of Bruno de Finetti. Stud. Bayesian Econometrics Statist, edited by Goel, P.K. & Zellner, A. North-Holland Publishing Co., Amsterdam. 6: 233-243.

Zheng, Y., Ooi, M.C.G., Juneng, L., Wee, H.B., Latif, M.T., Nadzir, M.S.M., Hanif, N.M., Chan, A., Li, L., Ahmad, N. & Tangang, F. 2023. Assessing the impacts of climate variables on long-term air quality trends in Peninsular Malaysia. Science of The Total Environment 901: 166430.

 

*Corresponding author; email: aftar@ukm.edu.my