Determination of the Best Probability Distribution of Fit for Ozone Concentration Data in Campo Grande-MS-Brazil

Abstract

This study discussed the behavior of ozone level observed in the atmospheric region of Campo Grande. To determine the best adjusted distribution to describe the ozone co-generation data for the year 2016 in Campo Grande were used 15 functions adjusted for this purpose; the performances of the distributions are evaluated using three test qualities, namely Kolmogorov-Smirnov, Anderson-Darling and Chi-Square test. Finally, the result of the fitted quality test is compared, it was observed that the generalized extreme value distribution provides a good fit for the whole year and the distributions Gamma 3P; lognormal 3P; weibull and Gamma 3P for the seasons of the year: winter, spring, summer, autumn, which are empirically proven to be the most appropriate distribution of data.