Forecasting Analysis of the Number of AIDS Incidence in China Based on the SARIMA-SVM Combination Model

Mengyao Du; Xiao Chang

doi:10.62051/ijphmr.v3n2.03

Authors

Mengyao Du
Xiao Chang

DOI:

https://doi.org/10.62051/ijphmr.v3n2.03

Keywords:

AIDS incidence number, SARIMA model, SARIMA-SVM combination model, Empirical analysis, Prediction

Abstract

Based on the monthly incidence data of AIDS, a SARIMA model and a SARIMA-SVM combination model are constructed to study and predict the number of cases, providing a scientific basis for the prevention and control of AIDS. The monthly incidence count data of AIDS from January 2004 to September 2023 are used as the training set, and the monthly incidence count data from October 2023 to September 2024 are used as the test set. Using R software, these two models are applied to predict the monthly incidence count of AIDS in China, and the predictions are compared with the actual values to evaluate the forecasting effectiveness of the single SARIMA model and the SARIMA-SVM hybrid model. The overall trend of AIDS incidence in China from 2004 to 2024 shows an upward trend, with certain seasonal characteristics. The root mean square error (RMSE) of the SARIMA model's forecast results is approximately 578.718, and the mean absolute percentage error (MAPE) is about 11.163%. The root mean square error (RMSE) of the SARIMA-SVM hybrid model's forecast results is approximately 45.643, and the mean absolute percentage error (MAPE) is about 0.838%. The SARIMA-SVM combination model has a good fit and prediction effect on the number of AIDS monthly incidence, and the prediction accuracy is high, which can be used to predict the month-by-month incidence data of AIDS.

References

[1] “The Joint United Nations Programme on HIV/AIDS (UNAIDS) has released a new report: In 2022, there were 1.3 million new HIV infections globally”. National Center for AIDS/STD Control and Prevention, China CDC. 2023. https://ncaids.chinacdc.cn/xxgx/yqbh/

[2] “In the Name of Love, Speaking for 'AIDS' —— A Data-Driven Review of 15 Years of AIDS in China”. The Paper. 2023. https://www.thepaper.cn/newsDetail_forward_25641879

[3] Yuxiu Guo. Analysis of AIDS Epidemic Characteristics and Prediction of Monthly Incidence Rate in China from 2004 to 2016 [D]. Anhui Medical University, 2019.

[4] Wang S, Wei F, Li H, Wang Z, Wei P. Comparison of SARIMA model and Holt-Winters model in predicting the incidence of Sjögren's syndrome. Int J Rheum Dis. 2022 Nov; 25(11):1263-1269. doi: 10.1111/1756-185X.14417. Epub 2022 Aug 12. PMID: 35962522.

[5] Wu WW, Li Q, Tian DC, Zhao H, Xia Y, Xiong Y, Su K, Tang WG, Chen X, Wang J, Qi L. Forecasting the monthly incidence of scarlet fever in Chongqing, China using the SARIMA model. Epidemiol Infect. 2022 Apr 21; 150:e90. doi: 10.1017/S0950268822000693. PMID: 35543101; PMCID: PMC9102071.

[6] Cong J, Ren M, Xie S, Wang P. Predicting Seasonal Influenza Based on SARIMA Model, in Mainland China from 2005 to 2018. International Journal of Environmental Research and Public Health. 2019; 16(23):4760.

[7] Tohidinik, H.R., Mohebali, M., Mansournia, M.A., Niakan Kalhori, S.R., Ali-Akbarpour, M. and Yazdani, K. (2018), Forecasting zoonotic cutaneous leishmaniasis using meteorological factors in eastern Fars province, Iran: a SARIMA analysis. Trop Med Int Health, 23: 860-869.

[8] Yaya Yang, Jie Liang, Shukun Liu. Application of GM (1, 1) Model for Forecasting the Incidence Trend of AIDS in Shaanxi Province [J]. Modern Medicine and Health Research Electronic Journal, 2019, 3(21):141-143.

[9] Shunyong Li, Kexin Li. Application of ARIMA and Prophet Models in Predicting the Number of AIDS Cases [J]. Henan Science, 2020, 38(09):1387-1393.

[10] Kui Chen, Chenxue Dong, Jiayue Lu et al. Forecast of the Number of AIDS Patients in China Based on ARIMA Time Series Model [J]. Primary Health Care, 2022, 36(03):91-93.

[11] Sayeed, M.A., Rahman, A., Rahman, A. et al. On the interpretability of the SVM model for predicting infant mortality in Bangladesh. J Health Popul Nutr 43, 170 (2024).

[12] Li, Z., Li, Y. A comparative study on the prediction of the BP artificial neural network model and the ARIMA model in the incidence of AIDS. BMC Med Inform Decis Mak 20, 143 (2020).

[13] Zaijin Guo, Hao Gong, Yunshui Li et al. Application of ARIMA and ARIMA-SVM hybrid model in the prediction of HIV incidence in Jiangsu [J]. Chinese Preventive Medicine, 2023, 24(08): 857-860. DOI:10.16506/j.1009-6639.2023.08.017.

[14] Lan Ma, Wenwen Yang, Xiaoling Ma et al. Application of SARIMA model and Holt-Winters exponential smoothing model in HIV/AIDS case prediction in Xinjiang Production and Construction Corps [J/OL]. Journal of Preventive Medicine Information:1-7 [2024-06-21]. https://doi.org/10.19971/j.cnki.1006-4028.230532.

[15] Mengmeng Li, Jinghui Zhang. Research Progress on the Status Quo of AIDS Epidemic [J]. Advances in Clinical Medicine, 2023, 13(2): 1601-1606.

[16] Chaoyang Guo, Huiying Liu, Shengchao Liu et al. Application of SARIMA-SVM combination model in the prediction of gonorrhea incidence [J]. Modern Disease Control and Prevention, 2024, 35(02):113-117. DOI:10.13515/j.cnki.hnjpm.1006-8414.2024.02.011.

[17] Yu Zhao, Suisui Chen, Liping Lv. Application of SARIMA-SVM Model in Tuberculosis Incidence [J]. Mathematics in Practice and Theory, 2023, 53(09):133-141.

[18] Fan Zhang, Luning Xun, Jixin Sun et al. Mortality prediction of road traffic injury based on ARIM-SVM combination model [J]. Modern Preventive Medicine, 2021, 48(10):1742-1746.