Multiple Linear Regression (MLR) and Principal Component Regression (PCR) for Ozone (O3) Concentrations Prediction

Abstract Rapid economic growth has led to an increase in ozone (O3) concentration which significantly affecting human health and environment. The prediction of O3 is complicated due to the redundancy of influencing parameters which introduce the multicollinearity problem. The aim of this study is to assess the best prediction model for O3 concentration which is Multiple Linear Regression (MLR) and Principle Component Regression (PCR). Data from 2012 to 2014 were used including O3, nitrogen dioxide (NO2), nitrogen oxide (O2), temperature, relative humidity and wind speed on hourly basis. Principle Component Analysis (PCA) was used in order to reduce multicollinearity problem, prior to the implementation of MLR. The hybrid model of PCR was selected as best -fitted models as it had higher correlation coefficient, R2 values compared with MLR model. In conclusion, the information from best-fitted prediction model can be used by local authorities to plan the precaution measure in combating and preserve the better air quality level.