Forecasting Non-Gaussian Time Series with TB Data
DOI:
https://doi.org/10.22399/ijcesen.3446Keywords:
Non-Gaussian, Gamma-ARIMA, EGWO algorithm, Bayesian inference, Tuberculosis, IraqAbstract
Abstract
Conventional forecasting models require time series that are stationary over time in terms of mean and
variance. However, we often encounter data that rarely meet this condition. The data may have Non-
Gaussian (N-G) distribution or contain heavy tails or extreme values. In order to improve and strengthen
the predictive performance, various (N-G) models have been used, each of which has a different property
from the other models. The combined formulas of discrete distributions such as Poisson or Negative –
Binomial (NB) distribution with Autoregressive Integrated Moving Average (ARIMA) models provide an
interpretable methodology when modeling time series data by following the characteristics of count data
because it relies on the distributional properties represented by the general linear model based on count
data and the time dependence represented by the ARIMA model of the residuals. Predicting time-
dependent patterns of count data involves complexities resulting from the discrete and positive nature of
the data, which is not compatible with the classical ARIMA methodology. To address this shortcoming,
models combining the two were used as an alternative solution. These models are Gamma-ARIMA,
Poisson-ARIMA, and NB- ARIMA. To fit discrete data to a continuous gamma distribution, a new
framework, the transformed Gamma-ARIMA model, was proposed. By applying a mathematical
transformation to discrete data, the series formation becomes more consistent, and the Gamma-ARIMA
technique is successful on non-Gaussian discrete data sets.. Four different mathematical formulations
were used, and the Enhanced Grey Wolf Optimizer (EGWO) algorithm was used to compare them. The
results show that the square root transformation is the best using the No-U-Turn Sampler (NUTS)
algorithm, and that the Bayesian estimation performance is robust and suitable for reliable inference and
future predictions. Using an annual time series of the number of pulmonary Tuberculosis (TB) cases in
Iraq, the results showed that the Poisson-ARIMA model outperformed the other models using Mean
Square Error (MSE)and Mean Absolute Percentage Error (MAPE).
Keywords: Non-Gaussian; Gamma-ARIMA; EGWO algorithm; Bayesian inference; Tuberculosis; Iraq.
References
[1] Balakrishna, N. (2021). Non-Gaussian autoregressive-type time series. Springer. DOI: https://doi.org/10.1007/978-981-16-8162-2
[2] Box, G. E. P., Jenkins, G. M., & Reinsel, G. C. (2009). Time series analysis: Forecasting and control (3rd ed.). Pearson. DOI: https://doi.org/10.1002/9781118619193.ch5
[3] Wang, J., & Chen, H. (2024). A bi-Gamma distribution model for a broadband non-Gaussian random stress rainflow range based on a neural network. Applied Sciences, 14(18), 8376. https://doi.org/10.3390/app14188376 DOI: https://doi.org/10.3390/app14188376
[4] Ng, J., Forbes, C. S., Martin, G. M., & McCabe, B. P. M. (2013). Non-parametric estimation of forecast distributions in non-Gaussian, non-linear state space models. International Journal of Forecasting, 29(3), 411–430. https://doi.org/10.1016/j.ijforecast.2012.10.005 DOI: https://doi.org/10.1016/j.ijforecast.2012.10.005
[5] Kaur, S., & Rakshit, M. (2019). Gaussian and non-Gaussian autoregressive time series models with rainfall data. International Journal of Engineering and Advanced Technology, 9(1). DOI: https://doi.org/10.35940/ijeat.A1994.109119
[6] Szarek, D., Jabłoński, I., Zimroz, R., & Wyłomańska, A. (2023). Non-Gaussian feature distribution forecasting based on ConvLSTM neural network and its application to robust machine condition prognosis. Expert Systems with Applications, 230, 120588. https://doi.org/10.1016/j.eswa.2023.120588 DOI: https://doi.org/10.1016/j.eswa.2023.120588
[7] Al-Mudhaffar, M. A. S. (1979). Spatial analysis of endemic diseases in Iraq: A study of the geographical foundations of health planning. Al-Irshad Press.
[8] Ali, A. H. (2007). The impact of administrative functions on the DOTS program for the treatment of tuberculosis in Iraq (Master’s thesis). St. Clements International University.
[9] Ministry of Health – Department of Public Health and Primary Health Care, Institute of Chest and Respiratory Diseases. (2023). Epidemiology of tuberculosis in Iraq.
[10] Kilicman, A., & Roslan, U. A. M. (2009). Tuberculosis in the Terengganu region: Forecast and data analysis. ScienceAsia, 35, 392–395. DOI: https://doi.org/10.2306/scienceasia1513-1874.2009.35.392
[11] Moosazadeh, M., Nasehi, M., Bahrampour, A., Khanjani, N., Sharafi, S., & Ahmadi, S. (2014). Forecasting tuberculosis incidence in Iran using Box-Jenkins models. Iran Red Crescent Medical Journal, 16(5), e11779. DOI: https://doi.org/10.5812/ircmj.11779
[12] Wang, K. W., Deng, C., Li, J. P., Zhang, Y. Y., Li, X. Y., & Wu, M. C. (2017). Hybrid methodology for tuberculosis incidence time-series forecasting based on ARIMA and a NAR neural network. Epidemiology and Infection, 145(6), 1118–1129. https://doi.org/10.1017/S0950268816003216 DOI: https://doi.org/10.1017/S0950268816003216
[13] Keerqinfu, Zhang, Q., Yan, L., & He, J. (2018). Time series analysis of correlativity between pulmonary tuberculosis and seasonal meteorological factors based on theory of Human-Environmental Inter Relation. Journal of Traditional Chinese Medical Sciences, 5(2), 119–127. https://doi.org/10.1016/j.jtcms.2018.03.001 DOI: https://doi.org/10.1016/j.jtcms.2018.03.001
[14] Wang, H., Tian, C. W., Wang, W. M., & Luo, X. M. (2018). Time-series analysis of tuberculosis from 2005 to 2017 in China. Epidemiology and Infection, 146, 935–939. https://doi.org/10.1017/S0950268818001115 DOI: https://doi.org/10.1017/S0950268818001115
[15] Sumi, A., & Kobayashi, N. (2019). Time-series analysis of geographically specific monthly number of newly registered cases of active tuberculosis in Japan. PLoS ONE, 14(3), e0213856. https://doi.org/10.1371/journal.pone.0213856 DOI: https://doi.org/10.1371/journal.pone.0213856
[16] Yang, C., et al. (2020). Analysis and prediction of pulmonary tuberculosis using an ARIMA model in Shaanxi Province, China. Journal of Physics: Conference Series, 1624, 022013. DOI: https://doi.org/10.1088/1742-6596/1624/2/022013
[17] Olanrewaju, S. O., Ojo, E. O., & Oguntade, E. S. (2020). Time series analysis on reported cases of tuberculosis in Minna Niger State Nigeria. Open Journal of Statistics, 10, 412–430. https://doi.org/10.4236/ojs.2020.103027 DOI: https://doi.org/10.4236/ojs.2020.103027
[18] Zheng, Y., Zhang, X., Wang, X., et al. (2021). Predictive study of tuberculosis incidence by time series method and Elman neural network in Kashgar, China. BMJ Open, 11, e041040. https://doi.org/10.1136/bmjopen-2020-041040 DOI: https://doi.org/10.1136/bmjopen-2020-041040
[19] Orjuela-Cañón, A. D., Jutinico, A. L., Duarte González, M. E., Awad García, C. E., Vergara, E., & Palencia, M. A. (2022). Time series forecasting for tuberculosis incidence employing neural network models. Heliyon, 8(7), e09897. https://doi.org/10.1016/j.heliyon.2022.e09897 DOI: https://doi.org/10.1016/j.heliyon.2022.e09897
[20] Wyld, D. C., et al. (2022). Towards tuberculosis incidence time series forecasting in Colombian regions: The Antioquia case. Computer Science & Information Technology, 117–126.
[21] Siamba, S., Otieno, A., & Koech, J. (2023). Application of ARIMA, and hybrid ARIMA models in predicting and forecasting tuberculosis incidences among children in Homa Bay and Turkana Counties, Kenya. PLOS Digital Health, 2(2), e0000084. https://doi.org/10.1371/journal.pdig.0000084 DOI: https://doi.org/10.1371/journal.pdig.0000084
[22] Ab Rashid, M. A., Ahmad Zaki, R., Wan Mahiyuddin, W. R., & Yahya, A. (2023). Forecasting new tuberculosis cases in Malaysia: A time-series study using the autoregressive integrated moving average (ARIMA) model. Cureus, 15(9), e44676. https://doi.org/10.7759/cureus.44676 DOI: https://doi.org/10.7759/cureus.44676
[23] Yerdessov, S., Arupzhanov, I., Aimyshev, T., Makhammajanov, Z., Kadyrov, S., Kashkynbayev, A., & Gaipov, A. (2023). Time series analysis on tuberculosis cases in Kazakhstan from 2014 to 2025. European Journal of Public Health, 33(Suppl_2). https://doi.org/10.1093/eurpub/ckad160.888 DOI: https://doi.org/10.1093/eurpub/ckad160.888
[24] Yadav, B. K., Srivastava, S. K., Thillai Arasu, P., & Singh, P. (2023). Time series modeling of tuberculosis cases in India from 2017 to 2022 based on the SARIMA-NNAR hybrid model. Canadian Journal of Infectious Diseases and Medical Microbiology, Article ID 5934552. https://doi.org/10.1155/2023/5934552 DOI: https://doi.org/10.1155/2023/5934552
[25] Sumanto, Y. D., Hariyanto, S., & Andriyati, A. (2024). Forecasting TB in Indonesia using ARMA(1,1). International Journal of Software & Hardware Research in Engineering, 12(9), 8–13. DOI: https://doi.org/10.26821/IJSHRE.12.9.2024.120905
[26] Elhag, A. A. (2024). Prediction and classification of tuberculosis using machine learning. Journal of Statistics Applications & Probability, 13(3), 939–946. DOI: https://doi.org/10.18576/jsap/130308
[27] Silva, M. T., & Galvão, T. F. (2024). Tuberculosis incidence in Brazil: Time series analysis between 2001 and 2021 and projection until 2030. Revista Brasileira de Epidemiologia, 27, e240027. https://doi.org/10.1590/1980-549720240027 DOI: https://doi.org/10.1590/1980-549720240027.2
[28] Lv, Z., Sun, R., Liu, X., et al. (2024). Evaluating the effectiveness of self-attention mechanism in tuberculosis time series forecasting. BMC Infectious Diseases, 24, 1377. https://doi.org/10.1186/s12879-024-10183-9 DOI: https://doi.org/10.1186/s12879-024-10183-9
[29] Maipan-Uku, J. Y., & Cavus, N. (2024). Forecasting tuberculosis incidence: A review of time series and machine learning models for prediction and eradication strategies. International Journal of Environmental Health Research. Advance online publication. https://doi.org/10.1080/09603123.2024.2368137 DOI: https://doi.org/10.1080/09603123.2024.2368137
[30] Chang, M., Emam, M., Zhang, N., et al. (2025). Analysis of changes in tuberculosis incidence trends before and after COVID-19 based on time series models. Advances in Continuous and Discrete Models, 48. https://doi.org/10.1186/s13662-025-03905-z DOI: https://doi.org/10.1186/s13662-025-03905-z
[31] Żuławiński, W., Maraj-Zygmąt, K., Shiri, H., Wyłomańska, A., & Zimroz, R. (2023). Framework for stochastic modelling of long-term non-homogeneous data with non-Gaussian characteristics for machine condition prognosis. Mechanical Systems and Signal Processing, 184, 109677. DOI: https://doi.org/10.1016/j.ymssp.2022.109677
[32] Azizi, A. (2014). Integration of seasonal autoregressive integrated moving average and Bayesian methods to predict production throughput under random variables. Journal of Mechanical Engineering and Sciences, 7, 1236–1250. https://doi.org/10.15282/jmes.7.2014.23.0121 DOI: https://doi.org/10.15282/jmes.7.2014.23.0121
[33] Vehtari, A., Gelman, A., & Gabry, J. (2017). Practical Bayesian model evaluation using leave-one-out cross-validation and WAIC. Statistics and Computing, 27, 1413–1432. https://doi.org/10.1007/s11222-016-9696-4 DOI: https://doi.org/10.1007/s11222-016-9696-4
[34] Helske, J., & Vihola, M. (2021). bssm: Bayesian inference of non-linear and non-Gaussian state space models in R. The R Journal, 13(2). DOI: https://doi.org/10.32614/RJ-2021-103
[35] Rajakumar, R., Sekaran, K., Hsu, C.-H., & Kadry, S. (2021). Accelerated grey wolf optimization for global optimization problems. Technological Forecasting and Social Change, 169, 120824. https://doi.org/10.1016/j.techfore.2021.120824 DOI: https://doi.org/10.1016/j.techfore.2021.120824
[36] Mirjalili, S., Mirjalili, S. M., & Lewis, A. (2014). Grey wolf optimizer. Advances in Engineering Software, 69, 46–61. https://doi.org/10.1016/j.advengsoft.2013.12.007 DOI: https://doi.org/10.1016/j.advengsoft.2013.12.007
[37] Pala, L. O. D. E., Carvalho, M. D. M., & Sáfadi, T. (2023). Analysis of count time series: A Bayesian GARMA (p, q) approach. Austrian Journal of Statistics, 52, 131–151. https://doi.org/10.17713/ajs.v52i5.1568 DOI: https://doi.org/10.17713/ajs.v52i5.1568
[38] Jia, Y. (2018). Some models for count time series (Doctoral dissertation). Clemson University. https://open.clemson.edu/all_dissertations/2213
[39] Razali, N. M., & Wah, Y. B. (2011). Power comparisons of Shapiro-Wilk, Kolmogorov-Smirnov, Lilliefors and Anderson-Darling tests. Journal of Statistical Modeling and Analytics, 2(1), 21–33.
[40] Wazeer, E. A. A., & Hameed, L. M. A. (2021). Forecasting the exchange rates of the US dollar against the Iraqi dinar using the Box-Jenkins methodology in time series with practical application. International Journal of Agricultural and Statistical Sciences, 17, 2085. https://connectjournals.com/03899.2021.17.2085
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2025 International Journal of Computational and Experimental Science and Engineering
This work is licensed under a Creative Commons Attribution 4.0 International License.
