Empirical Evaluation of Wavelet Filter and Wavelet Decomposition Level on Time Series Forecasting

Mira Andriyani, Dewi Retno Sari S.

Abstract


Time series forecasting is essential for anticipating future outcomes and supporting decision-making, yet achieving high predictive accuracy remains challenging. Wavelet-based approaches, particularly the Maximal Overlap Discrete Wavelet Transform (MODWT), offer potential improvements, although limited studies have systematically compared wavelet filter types and decomposition levels. This study evaluates several wavelet filters and decomposition levels combined with ARIMA models across six datasets exhibiting varying temporal characteristics. Forecasting accuracy was measured using the Mean Absolute Error (MAE) and Symmetric Mean Absolute Percentage Error (SMAPE). For the datasets analyzed, the Haar filter yielded the lowest MAE and SMAPE values, a result supported by the Kruskal–Wallis test and Dunn's test, which indicated significant differences in accuracy across filters. In contrast, differences in decomposition levels were not statistically significant, suggesting that decomposition level played a limited role in forecasting performance within this dataset context. These findings provide empirical, dataset-specific evidence regarding filter selection in MODWT–ARIMA modeling and highlight the comparatively minor influence of decomposition level on forecasting accuracy.

Keywords


ARIMA; level; MODWT; Wavelet Filter

Full Text:

PDF

References


[1] W. W. Wei, Time Series Analysis Univariate and Multivariate Methods. Pearson Education, Inc., 2006, vol. 2.

[2] S. Adamala, “Time series analysis: A hydrological prospective,” American Journal of Scientific Research and Essays, vol. 1, no. 1, pp. 31–40, 2016.

[3] E. Aladag, “Forecasting of particulate matter with a hybrid ARIMA model based on wavelet transformation and seasonal adjustment,” Urban Climate, vol. 39, pp. 100930–100945, 2021. doi: 10.1016/j.uclim.2021.100930.

[4] J. Bruzda, “The haar wavelet transfer function model and its applications,” Dynamic Econometric Models, vol. 11, pp. 141–153, 2011. doi: 10.12775/DEM.2011.010.

[5] H. P. M., M. Z. Rehman, A. A. Dar, and T. W. A., “Forecasting co2 emissions in India: A time series analysis using ARIMA,” Processes, vol. 12, no. 12, pp. 2699–2714, 2024. doi: 10.3390/pr12122699.

[6] S. Al Wadi, A. Hamarsheh, and H. Alwadi, “Maximum overlapping discrete wavelet transform in forecasting banking sector,” Applied Mathematical Sciences, vol. 7, no. 80, pp. 3995–4002, 2013. doi: 10.12988/ams.2013.36305.

[7] N. A. Yaacob, J. J. Jaber, D. Pathmanathan, S. Alwadi, and I. Mohamed, “Hybrid of the Lee–Carter model with maximum overlap discrete wavelet transform filters in forecasting mortality rates,” Mathematics, vol. 9, no. 18, pp. 2295–2305, 2021. doi: 10.3390/math9182295.

[8] M. U. Yousuf, I. Al-Bahadly, and E. Avci, “Short-term wind speed forecasting based on hybrid MODWT–ARIMA–Markov model,” IEEE Access, vol. 9, pp. 4803–4820, 2021. doi: 10.1109/ACCESS.2021.3084536.

[9] A. Jierula, S. Wang, T.-M. Oh, and P. Wang, “Study on accuracy metrics for evaluating the predictions of damage locations in deep piles using artificial neural networks with acoustic emission data,” Applied Science, vol. 11, p. 2314, 2021. doi: 10.3390/app11052314.

[10] L. Zhu, Y. Wang, and Q. Fan, “MODWT–ARMA model for time series prediction,” Applied Mathematical Modelling, vol. 38, no. 5–6, pp. 1859–1865, 2014. doi: 10.1016/j.apm.2013.10.002.

[11] K. Szostek, D. Mazur, G. Dralus, and J. Kusznier, “Analysis of the effectiveness of ARIMA, SARIMA, and SVR models in time series forecasting: A case study of wind farm energy production,” Energies, vol. 17, p. 4803, 2024. doi: 10.3390/en17194803.

[12] T. Ndlovu and D. Chikobvu, “A wavelet-decomposed WD–ARMA–GARCH–EVT model approach to comparing the riskiness of the Bitcoin and South African Rand exchange rates,” Data, vol. 8, no. 7, p. 122, 2023. doi: 10.3390/data8070122.

[13] A. A. A. Dghais and M. T. Ismail, “A study of stationarity in time series by using wavelet transform,” Proceedings of the 21st National Symposium on Mathematical Sciences (SKSM21), vol. 1605, pp. 798–804, 2014. doi: 10.1063/1.4887692.

[14] T. S. Alshammari, M. T. Ismail, S. Al-Wadi, M. H. Saleh, and J. J. Jaber, “Modeling and forecasting Saudi stock market volatility using wavelet methods,” Journal of Asian Finance, Economics and Business, vol. 7, no. 11, pp. 83–93, 2020. doi: 10.13106/jafeb.2020.vol7.no11.08.

[15] L.-W. Lin and X.-H. Zhou, “Multiscale forecasting approach of property insurance income via wavelet method,” Mathematical Problems in Engineering, vol. 2022, p. 9554695, 2022. doi: 10.1155/2022/9554695.

[16] P. Mittal, “Wavelet transformation and predictability of gold price index series with ARMA model,” International Journal of Experimental Research and Review, vol. 30, pp. 127–133, 2023. doi: 10.52756/ijerr.2023.v30.014.

[17] P. Mittal, “Forecasting of crude oil prices using wavelet decomposition based denoising with ARMA model,” Asia Pacific Financial Markets, vol. 31, pp. 355–365, 2024. doi: 10.1007/s10690-023-09418-7.

[18] J. Quilty and J. Adamowski, “A maximal overlap discrete wavelet packet transform integrated approach for rainfall forecasting: A case study in the Awash River Basin (Ethiopia),” Environmental Modelling and Software, vol. 144, pp. 105119–105133, 2021. doi: 10.1016/j.envsoft.2021.105119.

[19] G. Chiranjivi and R. Sensarma, “The effects of economic and financial shocks on private investment: A wavelet study of return and volatility spillovers,” International Review of Financial Analysis, vol. 90, p. 102936, 2023. doi: 10.1016/j.irfa.2023.102936.

[20] S. Al Wadi, O. Al Singlawi, J. J. Jaber, M. H. Saleh, and A. A. Shehadeh, “Enhancing predictive accuracy through the analysis of banking time series: A case study from the Amman Stock Exchange,” Journal of Risk and Financial Management, vol. 17, no. 3, p. 98, 2024. doi: 10.3390/jrfm17030098.

[21] H. Farajpanah et al., “A novel application of waveform matching algorithm for improving monthly runoff forecasting using wavelet–ML models,” Journal of Hydroinformatics, vol. 26, no. 7, pp. 1771–1789, 2024. doi: 10.2166/hydro.2024.128.

[22] M. Panja, T. Chakraborty, U. Kumar, and N. Liu, “Epicasting: An ensemble wavelet neural network for forecasting epidemics,” Neural Networks, vol. 165, pp. 185–212, 2023. doi: 10.1016/j.neunet.2023.05.049.

[23] Hermansah, D. Rosadi, H. Utami, Abdurakhman, and G. Darmawan, “Hybrid MODWT–FFNN model for time series data forecasting,” AIP Conference Proceedings, vol. 2192, 2019. doi: 10.1063/1.5139175.

[24] P. J. Brockwell and R. A. Davis, Introduction to Time Series and Forecasting. Springer, 2016. doi: 10.1007/978-3-319-29854-2.

[25] Badan Pusat Statistik, “Jumlah penumpang dan barang melalui transportasi kereta api Indonesia tahun 1987–2022.” Available: https://www.bps.go.id/id/statistics-table/1/MTQxNCMx/jumlah-penumpang-dan-barang-melalui-transportasi-kereta-api-indonesia-tahun-1987-2022.html. Accessed: Jul. 14, 2025.

[26] A. D. Kayit and M. T. Ismail, “Advancing stock price prediction through the development of hybrid ensembles: A comprehensive comparative analysis of machine learning approaches,” Journal of Big Data, vol. 12, p. 232, 2025. doi: 10.1186/s40537-025-01185-8.

[27] D. Ramadhani, A. M. Soleh, and Erfiani, “Characteristics of machine-learning-based univariate time series imputation method,” JUITA: Jurnal Informatika, vol. 12, pp. 279–288, 2024. doi: 10.30595/juita.v12i2.23453.

[28] A. Dinno, “Nonparametric pairwise multiple comparisons in independent groups using Dunn’s test,” The Stata Journal, vol. 15, pp. 292–300, 2015.




DOI: https://doi.org/10.18860/cauchy.v10i2.36440

Refbacks

  • There are currently no refbacks.


Copyright (c) 2025 Mira Andriyani, Dewi Retno Sari S.

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

Editorial Office
Mathematics Department,
Universitas Islam Negeri Maulana Malik Ibrahim Malang
Gajayana Street 50 Malang, East Java, Indonesia 65144
Faximile (+62) 341 558933
e-mail: cauchy@uin-malang.ac.id

Creative Commons License
CAUCHY: Jurnal Matematika Murni dan Aplikasi is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.