Hockey Stick回归和SARIMA模型在细菌性痢疾疫情预测中的应用  被引量:6

Prediction of Incidence of Bacillary Dysentery Based on Hockey Stick Regression and SARIMA Model

在线阅读下载全文

作  者:关鹏[1] 王汉宁[2] 沈铁峰[3] 吴伟[1] 黄德生[4] 周宝森[1] 

机构地区:[1]中国医科大学公共卫生学院流行病学教研室,辽宁沈阳110001 [2]中国医科大学 [3]辽宁省葫芦岛市疾病预防控制中心 [4]中国医科大学基础医学院数学教研室

出  处:《实用预防医学》2008年第6期1671-1675,共5页Practical Preventive Medicine

基  金:国家自然科学基金资助项目(70503028);教育部留学回国人员科研启动基金资助项目(教外司留[2008]890号)

摘  要:目的探讨应用Hockey Stick回归和时间序列分析中的SARI MA模型进行细菌性痢疾疫情预测的可行性。方法收集辽宁省葫芦岛市1990-2006年的逐月及逐年细菌性痢疾疫情资料和当地气象数据。首先,利用描述统计分析细菌性痢疾季节性发病规律,使用Spearman等级相关分析疫情同气象因素的关系,根据Hockey Stick回归确定发病阈值。其次,进行扩充迪基富勒的平稳性单位根检验。再次,根据自相关函数图和偏自相关函数图识别逐月疫情间的相关性。应用Eviews3.1、Stata8.2和SPSS12.0软件对1990-2005年逐月发病率进行上述统计分析。最后,利用所得到的模型对2006年各月发病率进行预测,并与实际发病率进行比较。结果最低气温、平均气温和最高气温所确定的阈值分别为11.42℃、17.17℃和22.98℃;(1,0,0)×(0,1,1)12模型为最优SARI MA模型,此模型对2006年各月发病率的预测值符合实际发病率变动趋势。结论Hockey Stick回归和SARI MA模型可较好地模拟细菌性痢疾疫情在时间序列上的变动趋势,并对未来的发病率进行一定预测,能够为传染病防制工作提供一定决策支持。Objective To explore the feasibility of Hockey Stick. regression(HSR) and seasonal autoregressive integrated moving average(SARIMA) model to predict the incidence of bacillary dysentery(BD) in Huludao City. Methods Data on BD were provided by Huludao Municipa~ Center for Disease Control and Prevention, and meteorological data were re- trieved from the China Bureau of Meteorology. Firstly, tabular and pictorial descriptive statistics were used to analyze the seasonal rule of BD. Spearman rank correlation was performed for seeking the relationship between incidence of BD and climate variables. Climate thresholds were then determined by HSR. Secondly, augmented dickey fuller (ADF) unit root test was applied for series trend analysis. Thirdly, autocorrelation function (ACF) and partial autocorrelation function (PACF) graphs were used to detect auto- correlationship of the BD incidence. Eviews (version 3.1), Stata (version 8.2) and SPSS (version 12.0) softwares were used to perform the above statistical analyses based on the monthly incidence of BD in Huludao from January, 1990 to December, 2005. Finally, the constructed model was used to predict the monthly incidence in 2006 and the prediction values were compared with actual incidences. Results Thresholds for minimum temperature, average temperature, and maximum temperature were 11.42℃, 17.17℃ and 22.98 ℃, respectively. SARIMA(1,0,0) ×(0,1, 1)12 exactly fitted the incidences of the previous months, and incidences in 2005 estimated by the model were consistent with the a tual values. Conclusions The methods of HSR and SARIMA can be used to fit the changes of the incidence of BD and to predict the future incidence.

关 键 词:流行病学 预测 HOCKEY Stick回归 SARIMA 细菌性痢疾 

分 类 号:R195.1[医药卫生—卫生统计学]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象