Constructing A Model to Study the Effect of Some Variables on Air Pollution in Baghdad Governorate for the Period (2017-2019)

Abstract

Pollution is one of the important problems facing humanity at the present time as a result of the increase in human activity in various fields of life, and there are different types of forms of pollution, and air pollution is one of the serious aspects of pollution, so a study was conducted using the logistic regression model, as it is considered one of the effective and appropriate models in The process of descriptive data analysis of the binary response, and the Probit model, which is similar to the binary logistic model in the nature of the dependent variable, because it is a qualitative variable that takes two properties, zero and one, and depends on the probability density function and the cumulative distribution function, and follows the standard normal distribution. The effect of influential variables that represent gases emitted into the air (CO, NO, CH4, SO2, NO2, NOx) on the response variable, which represents suspended particles within the local determinant of pollution within 24 hours (PM10, PM2.5), where the law of low pollution rate was adopted and elevated which takes (0,1) and this variable was converted into a binary response based on the measured proportions of this variable. The study took three stations in the Governorate of Baghdad, which are (Al-Waziriyah, Al-Andalus Square, and Al-Saidiah). The research found that the high level of pollution in suspended particles is due to the influencing variable (CH4) in Waziriya station and the two variables (SO2, NOx) in Meydan Al-Andalus station, and the variable (NO2, CH4) at Saidia station.