Ozone concentrations Predicting in Agadir city (Morocco) using the Multiple Linear Regression (Forward Regression Analysis)

Nowadays, Given the harm that air pollution causes to human health and the environment, it is becoming more and more vital to investigate and predict air quality indices. The ground level ozone is one of the most dangerous pollutants. The purpose of this study is to apply the forward regression to forecast the daily average ozone concentration in the city of Agadir (urban area), based on meteorological parameters and primary pollutants in order to investigate its contributions on ozone concentrations and assist policymakers in reducing the negative impacts that ozone has on human health. A mobile monitoring station was used to collect daily average data in 2013 for training and testing the statistical model. In order to evaluate the effectiveness of the forecast, we calculated some statistical score such as coefficient of correlation (r), coefficient of determination (R2), Mean Absolute Error (MAE), and Root Mean Square Error (RMSE). The obtained results indicated the method's capacity to forecast ozone concentrations with a reasonable scores values (r= 0.942, R2= 0.886, RMSE=1.258, MAE=0.902, MSE=1.584).