基于ARIMA模型的江苏省梅毒疫情预测

发布时间：2018-04-23 18:15

本文选题：ARIMA模型 + 时间序列分析　；参考：《南京医科大学学报(自然科学版)》2017年05期

【摘要】：目的:了解江苏省梅毒的流行病学特点,构建预测江苏省梅毒月发病率的自回归移动平均模型(autoregressive integrated moving average model,ARIMA),为梅毒预防控制工作提供参考依据。方法:利用江苏省1995—2009年梅毒月发病率资料建立ARIMA预测模型,并进行模型评价。结果:拟合ARIMA(1,1,0),(2,1,0)模型为预测江苏省梅毒月发病率的最佳模型,模型周期性、季节性一阶、二阶系数分别为-0.579、-0.245、-0.357,t检验统计量分别为8.777,2.881,4.766,相应的P值分别为0.001、0.005、0.001,表明该模型具有较高的预测精度,预测值与实际值较为接近,且实际值均在预测值的95%置信区间范围内,预测效果较好。结论:ARIMA模型能较好地预测梅毒发病率的变化趋势,为梅毒预防控制措施的制定提供重要依据。
[Abstract]:Objective: to investigate the epidemiological characteristics of syphilis in Jiangsu Province, and to construct an autoregressive integrated moving average model to predict the monthly incidence of syphilis in Jiangsu Province, so as to provide a reference for the prevention and control of syphilis. Methods: based on the monthly syphilis incidence data from 1995 to 2009 in Jiangsu Province, a ARIMA prediction model was established and evaluated. Results: the fitting model of Arima is the best model for predicting the monthly incidence of syphilis in Jiangsu Province. The model is periodic and seasonal. The second order coefficients are -0.579- 0.245- 0.357t test statistics are 8.777U 2.881g / 4.766and the corresponding P values are 0.001n0.00050.001respectively. The results show that the model has a high prediction accuracy, the predicted values are close to the actual values, and the actual values are within the 95% confidence interval of the predicted values. The prediction effect is good. Conclusion the weight Arima model can predict the trend of syphilis incidence and provide important basis for the establishment of syphilis prevention and control measures.
【作者单位】：南京医科大学公共卫生学院流行病与卫生统计学系;江苏省疾病预防控制中心;
【基金】：国家自然科学基金(81673275,U1503123) 国家“十二五”重大科技专项(2012ZX10001-001) 江苏省高校优势学科建设工程资助
【分类号】：R181.8;R759.1

【相似文献】