自回归求和移动平均乘积季节模型在西安地区出生缺陷预测中的应用

发布时间：2018-05-18 20:32

本文选题：出生缺陷 + 自回归求和移动平均乘积季节模型　；参考：《西安交通大学学报(医学版)》2017年03期

【摘要】：目的应用自回归求和移动平均(ARIMA)乘积季节模型预测西安市出生缺陷的发生率。方法利用2009年10月至2015年8月出生缺陷监测数据对西安市出生缺陷发生率数据构建ARIMA乘积季节模型,同时利用2015年9月至12月实际出生缺陷发生率与模型拟合数据进行比较,评价模型的预测性能,并预测西安市2016年的出生缺陷发生率。结果西安市出生缺陷的发生率具有一定的趋势及季节性,建立了ARIMA(0,0,1)(0,1,1)12乘积季节模型,利用2015年9月至12月拟合值与实际出生缺陷发生率比较,绝对误差的平均9.5,相对误差的平均0.084,提示ARIMA(0,0,1)(0,1,1)12乘积季节模型具有较佳的预测能力。预测2016年西安市出生缺陷发生率与2015年接近,总体略有抬升,但峰值下降。结论 ARIMA(0,0,1)(0,1,1)12乘积季节模型可用于西安市出生缺陷发生率的预测。
[Abstract]:Objective to predict the incidence of birth defects in Xi'an by using autoregressive and moving average Arima (ARIMA) product seasonal model. Methods using the birth defect monitoring data from October 2009 to August 2015, the seasonal model of ARIMA product was constructed, and the actual birth defect rate from September to December 2015 was compared with the fitting data. The predictive performance of the model was evaluated and the incidence of birth defects in Xi'an in 2016 was predicted. Results the incidence rate of birth defects in Xi'an city has a certain trend and seasonality. A seasonal model of Arima 0A0 / 0 / 0 / 1 / 1 / 1 / 1 / 12 product was established. The fitting value from September to December 2015 was used to compare the incidence of birth defects with that of actual birth defects. The average absolute error is 9.5, and the average relative error is 0.084, which indicates that the Arima model has a better prediction ability. The incidence of birth defects in Xi'an in 2016 is forecast to be close to 2015, with a slight increase in the overall level, but a decline in the peak. ConclusionThe seasonal model of Arima _ (1) ~ (1) ~ (1) ~ (1) ~ (1) ~ (1) ~ (1) ~ (1) ~ (12) product can be used to predict the incidence of birth defects in Xi'an.
【作者单位】：西安交通大学公共卫生学院流行病与卫生统计学系;西安市妇幼保健院;
【基金】：陕西省卫计委出生缺陷防治课题(No.sxwsjswzfcght2016-013) 国家自然科学基金资助项目(No.81230016)~~
【分类号】：R174

【相似文献】