Comparison of effects of ARIMA model and Holt-Winters exponential smoothing method in the prediction of suicide death
XUN Lu-ning1,2, CUI Ze2, SUN Ji-xin2, CAO Ya-jing2, ZHANG Fan2
1. School of Public Health, North China University of Science and Technology, Tangshan, Hebei 063210, China; 2. Hebei Provincial Center for Disease Control and Prevention, Shijiazhuang,Hebei 050011,China
Abstract：Objective To explore and compare the application of autoregressive integrated moving average (ARIMA) model and Holt-Winters exponential smoothing method in the prediction of suicide death. Methods The ARIMA model and the Holt-Winters exponential smoothing model were established respectively based on the data about the monthly death cases of suicide from January 2014 to June 2018 in Hebei Province. The monthly death cases of suicide from July to December 2018 were predicted and then compared with the actual number of deaths. And the prediction effects of the two models were evaluated based on root mean square error (RMSE), mean absolute error (MAE) and mean absolute percentage error (MAPE). Results A total of 2,882 suicide deaths were reported in Hebei Province from 2014 to 2018, and the overall level of suicide deaths showed a downward trend. The best ARIMA model established was ARIMA (0,1,1)(1,1,0)12, and RMSE, MAE and MAPE of the predicted results were 5.99, 4.67 and 9.80%, respectively. The best fitting model of Holt-Winters exponential smoothing method was the multiplicative model, and RMSE, MAE and MAPE of the predicted results were 6.03, 5.17 and 11.44%, respectively. Conclusion The prediction effect of ARIMA model is superior to that of Holt-Winters exponential smoothing method, and it is more suitable for short-term prediction of trend of suicide deaths.
寻鲁宁, 崔泽, 孙纪新, 曹亚景, 张帆. ARIMA模型和Holt-Winters指数平滑法在自杀死亡中的预测效果比较[J]. 实用预防医学, 2021, 28(6): 661-665.
XUN Lu-ning, CUI Ze, SUN Ji-xin, CAO Ya-jing, ZHANG Fan. Comparison of effects of ARIMA model and Holt-Winters exponential smoothing method in the prediction of suicide death. , 2021, 28(6): 661-665.
[1] World Health Organization. Preventing suicide: a resource for filmmakersand others working on stage and screen[R]. Geneva:WHO,2019:1-25. [2] Zhang L, Li Z, Li X, et al. Study on the trend and disease burden of injury deaths in Chinese population, 2004—2010[J]. PLoS One, 2014,9(1):e85319. [3] 杜睿,江光荣. 自杀行为:影响因素、理论模型及研究展望[J]. 心理科学进展,2015,23(8):1437-1452. [4] 郑名烺,陈辉. 麻城市自杀死亡水平时间序列模型分析及预测研究[J]. 中国卫生统计,2012,29(3):395-396. [5] 刘莹,胡锡敏,陈言,等. ARIMA模型在海南省2014—2017年居民自杀死亡率中的应用[J]. 中华流行病学杂志,2018,39(5):664-668. [6] 孟凡东,吴迪,隋承光. 2004—2015年中国狂犬病发病数据ARIMA乘积季节模型的建立及预测[J]. 中国卫生统计,2016,33(3):389-391,395. [7] 洪志敏,郝慧,房祥忠,等. ARIMA模型在京津冀区域手足口病发病趋势预测中的应用[J]. 数理统计与管理,2018,37(2):191-197. [8] 王燕. 时间序列分析—基于R[M]. 北京:中国人民大学出版社,2015:117-124. [9] 朱奕奕,赵琦,冯玮,等. 应用指数平滑法预测上海市甲型病毒性肝炎发病趋势[J].中国卫生统计,2013,30(1):31-33,36. [10] 朱佳佳,胡登利,洪秀琴,等. 基于时空大数据的甲型肝炎发病率分布特征分析及预测模型[J]. 中华疾病控制杂志,2018,22(11):1144-1147. [11] 汪鹏,彭颖,杨小兵. ARIMA模型与Holt-Winters指数平滑模型在武汉市流感样病例预测中的应用[J]. 现代预防医学,2018,45(3):385-389. [12] 宋媛媛,王雷,熊甜,等. ARIMA模型与GM(1,1)模型在痢疾发病数预测中的比较研究[J]. 实用预防医学,2019,26(7):888-892. [13] 严婧,杨北方. 指数平滑法与ARIMA模型在湖北省丙型病毒性肝炎发病预测中的应用[J]. 中国疫苗和免疫,2017,23(3):292-297. [14] 唐广心,张飞飞,鲁苇葭,等. 指数平滑法在麻疹发病率预测中的应用[J]. 实用预防医学,2018,25(6):757-759. [15] 高欣,金叶,汪媛,等.2006—2016年中国自杀死亡及自残/自杀病例流行特征分析[J]. 中华预防医学杂志,2019,53(9):885-890. [16] Deisenhammer EA, Stiglbauer C, Kemmler G. Time-related aspects of suicides - suicide frequency related to birthday, major holidays, day of the week, season, month of birth and zodiac signs[J]. Neuropsychiatr,2018,32(2):93-100. [17] 胡冬梅,姜潮,刘启贵,等. 应用集中度和圆形分布分析自杀的高发时间和季节[J]. 中国卫生统计,2004,21(6):360-361. [18] 高欣,王临虹,金叶,等. 1990年与2013年中国人群自杀疾病负担分析[J]. 中华流行病学杂志,2017,38(10):1325-1329.
2021, Vol. 28 
