Abstract:

Objective To explore the feasibility of autoregressive integrated moving average（ARIMA）to predict the infec? tion rates of hookworm in Jiangsu Province. Methods From 1990 to 2006，the infection rates of hookworm were used for a train? ing data set. As to obtain a stationary data set，the training data set was second?order differenced using the version SAS 9.0. The model parameters were screened by using the minimum information criterion. The ARIMA model was constructed to predict the in? fect rates of hookworm form 2007 to 2011. Results The time?series model ARIMA（1，2，0）was confirmed preliminarily. The model fitted well the training data set. The predictive infection rates were main accordance with the actual status of hookworm from 2007 to 2011，and the most minimum error was only 9.23%. Conclusion The model constructed has a good predictive effect and applied value for control of hookworm.

Key words: Hookworm； Time?series model； Autoregressive integrated moving average（ARIMA）； Jiangsu Province

摘要：

目的 探讨应用时间序列ARIMA模型预测江苏省钩虫感染率的可行性。方法 以1990-2006年江苏省钩虫感染率数据做为训练数据集， 应用SAS 9.0 软件对训练数据集进行差分平稳化处理后， 采用最小信息准则筛选参数， 构建全省钩虫病自回归滑动平均模型 （ARIMA）， 预测全省钩虫感染率。结果 初步确定全省钩虫感染率时间序列模型ARIMA （1， 2， 0）， 应用该模型预测的全省钩虫病流行趋势与实际感染情况相一致， 实际感染率均落在预测值95%可信区间内； 模型预测的2007-2011年全省钩虫感染率与实际感染率基本相符， 最小预测误差仅为9.23%。 结论 构建的时间序列模型具有良好的预测效果和一定的防治应用价值。

关键词: 钩虫； 时间序列模型； 自回归滑动平均模型； 江苏省