摘要
为了有效控制我国空气污染问题,改进空气质量预测方法,对特定污染物进行有针对性的监控和预防,不仅可以改善人们的生活质量,还有利于解决空气污染对各大城市发展的制约。本文将收集2018~2019年珠三角地区有代表性的8个城市的空气质量指标(AQI)的日数据,通过时间序列分析方法(ARIMA模型)对AQI数据进行拟合和分析,发现数据的变化趋势和规律,根据AIC、BIC最小准则对模型进行优化,并对未来五期的数据进行预测。结果表明:广州、深圳空气质量较好,两者都是珠三角地区的一线城市,珠海、江门、中山空气质量较差,空气质量与所在城市的发展规模有可能存在相关联系。通过预测结果和真实值对比,发现ARIMA模型预测精度高,拟合成功。
In order to effectively control the air pollution problem in China, improving air quality prediction methods and conducting targeted monitoring and prevention of specific pollutants can not only im-prove people’s quality of life, but also help solve the constraints of air pollution on the development of major cities. This article will collect daily data on air quality indicators (AQI) of 8 representative cities in the Pearl River Delta region from 2018 to 2019. The AQI data will be fitted and analyzed using time series analysis methods (ARIMA model) to discover the trends and patterns of data changes. Then we optimized the model based on the AIC and BIC minimum criteria, and predicted the data for the next five periods. The results indicate that Guangzhou and Shenzhen have good air quality, both of which are first tier cities in the Pearl River Delta region. Zhuhai, Jiangmen and Zhongshan have poor air quality, and there may be a correlation between air quality and the de-velopment scale of their respective cities. By comparing the predicted results with the actual values, it was found that the ARIMA model has high prediction accuracy and successful fitting.
出处
《统计学与应用》
2023年第5期1451-1463,共13页
Statistical and Application
