| 基于自回归滑动平均混合模型的布鲁菌病流行趋势预测 |
| |
| 引用本文： | 赵媛, 郭忠琴, 梁沛枫. 基于自回归滑动平均混合模型的布鲁菌病流行趋势预测[J]. 中华疾病控制杂志, 2019, 23(8): 932-937. doi: 10.16462/j.cnki.zhjbkz.2019.08.010 |
| |
| 作者姓名： | 赵媛 郭忠琴 梁沛枫 |
| |
| 作者单位： | 1.750001 银川, 宁夏医科大学公共卫生与管理学院流行病与卫生统计学系;;2.750004 银川, 宁夏回族自治区人民医院病案室 |
| |
| 基金项目： | 宁夏自然科学基金NZ17187西北民族大学中央高校基本科研业务费项目重点项目31920180088 |
| |
| 摘 要： | 目的 对我国布鲁菌病（简称布病）月发病率进行预测，为了解我国布病流行趋势、制定防控策略提供数据支持和决策依据。 方法 以国家人口与健康科学数据共享平台为数据来源，使用2004年1月-2016年12月全国布病月发病率数据建立历史序列，应用R软件构建自回归滑动平均混合模型（autoregressive integrated moving average model，ARIMA）并进行数据拟合和预测。 结果 本研究构建乘积季节ARIMA（2，1，2）（2，1，1）12模型各项参数都有统计学意义（均有P < 0.001），模型很好的拟合了全国布病月发病率的变化规律，预测值与实际值之间的平均相对误差为21.77%；预测2017年、2018年、2019年和2020年布病的月平均发病率分别为0.399 5/10万、0.423 8/10万、0.445 6/10万、0.471 2/10万，呈逐渐增高趋势（χ2=14.244，P < 0.001），在4-7月份出现发病峰值。 结论 在自然状况下，我国人间布病的月发病率将逐年增高，应采取相应措施进行控制。
|
| 关 键 词： | 布鲁菌病 时间序列 ARIMA模型 |
| 收稿时间： | 2019-03-12 |
| 修稿时间： | 2019-06-10 |
Prediction of brucellosis epidemic trend based on ARIMA model |
| |
| ZHAO Yuan, GUO Zhong-qin, LIANG Pei-feng. Prediction of brucellosis epidemic trend based on ARIMA model[J]. CHINESE JOURNAL OF DISEASE CONTROL & PREVENTION, 2019, 23(8): 932-937. doi: 10.16462/j.cnki.zhjbkz.2019.08.010 |
| |
| Authors: | ZHAO Yuan GUO Zhong-qin LIANG Pei-feng |
| |
| Affiliation: | 1. Department of Epidemiology and Biostatistics, School of Public Health and Management, Ningxia Medical University, Yinchuan 750001, China;;2. Case Room of People's Hospital of Ningxia Hui Autonomous Region, Yinchuan 750004, China |
| |
| Abstract: | Objective The aims is to predict the monthly incidence of brucellosis in China, in order to understand the epidemic trend of brucellosis in China, to formulate prevention and control strategies, and to provide data support and decision-making basis. Methods The national population and health science data sharing platform was used to collect the national incidence of brucellosis from January 2004 to December 2016 by month. The data were fitted and predicted using ARIMA model with R software. Results In this study, the parameters of the product season ARIMA (2, 1, 2) (2, 1, 1)12 model had statistical significance (all P < 0.001). The model fitted well the monthly incidence of brucellosis in China. The average relative error between the predicted value and the actual value was 21.77%. The monthly average incidence of brucellosis in 2017, 2018, 2019 and 2020 were predicted to be 0.399 5/100 000, 0.423 8/100 000, 0.445 6/100 000 and 0.471 2/100 000 respectively, showing a gradually increasing trend (χ2=14.244, P < 0.001), with a peak incidence from April to July. Conclusion Under natural conditions, the monthly incidence of human brucellosis in China will increase year by year, and corresponding measures should be taken to control it. |
| |
| Keywords: | Brucellosis Time series ARIMA model |
| 本文献已被 万方数据 等数据库收录！ |
| 点击此处可从《中华疾病控制杂志》浏览原始摘要信息 |
|
点击此处可从《中华疾病控制杂志》下载免费的PDF全文 |
|
