Analysis And Prediction Of Air Quality In Shandong Province Based On Quantile Regressio

As an essential "life gas" for our lives,the quality of air is always related to our health and living standards.However,as a large coal-fired province and a province with many industrial development cities,Shandong Province has long been dominated by coal energy consumption structure,together with its industrial pollution emissions,resulted in serious air pollution in the province,which not only affects people’s health,but also restricts the sustainable economic development of cities in Shandong Province.The emergence of heavy pollution and haze weather every year reminds us that we must attach great importance to the surrounding air quality,although a series of pollution control measures issued by the government in recent years have improved the overall air quality of Shandong Province to a certain extent.Therefore,it is more and more important to analyze and study the air quality of Shandong Province,predict its future trend,and then put forward feasible suggestions to improve its air quality and facilitate people’s daily travel.This paper applied quantile regression to the analysis of air quality,and collects the daily data of air quality in 10 representative cities of Shandong Province from 2018 to 2020,including air quality index AQI and PM2 5.PM10 and other six major air pollutants concentration,average temperature,average humidity and average wind speed.Firstly,descriptive statistical analysis of air quality is carried out on the three-year data of 10 cities;then established the panel quantile regression model,analyzed the influencing factors of AQI at different quantile levels and compared with the results of ordinary panel regression model;finally,used the collected variable data,established the quantile autoregressive distribution lag model to predict the change trend of AQI at different levels.The main conclusions are as follows:Firstly,in descriptive statistics,it is found that the air quality of coastal cities such as Weihai is good,while that of industrial cities such as Jinan and Zibo is poor.Moreover,the city closure and delayed resumption of work and production measures implemented in 2020 can improve the air quality of all cities to a certain extent under the influence of covid-19 pneumonia outbreak in late 2019.Secondly,through panel quantile regression modeling,it is found that the explanatory variables that have a significant impact on AQI at different quantiles are different,and the effect size and direction are also different.At each quantile,the first two variables that have the greatest impact on AQI are PM2 5 and PM10 are positive effects,and their promoting effect on AQI growth when the air quality is good is greater than that when the air quality is poor.On the contrary,the impact of CO on AQI when the air quality is poor is greater than that when the air quality is good,indicating that CO has played an important role in promoting the emergence of air pollution.In addition,the average wind speed is not the main influencing factor of AQI.Thirdly,through the trend prediction of AQI by quantile autoregressive distribution lag model,the approximate distribution range of AQI is obtained,and it is found that the prediction curve at 0.5 quantile is the closest to the real sequence,and the prediction effect is the best.The prediction curve of 0.975 quantile shows that under the condition of overall improvement of air quality,there will still be relatively serious pollution weather.Generally speaking,although the overall air quality of Shandong Province has a good development trend,it should not be too optimistic.To improve the air quality,we need to focus on controlling PM2 5 and PM10 concentrations.The above conclusions provide a theoretical basis for further policies to control air pollution in Shandong Province.