An assessment and prediction of PM_2.5 for a port city of eastern peninsular India is presented. Fifteen machine learning (ML) regression models were trained, tested and implemented to predict the PM_2.5 concentration. The predicting ability of regression models was validated using air pollutants and meteorological parameters as input variables collected from sites located at Visakhapatnam, a port city on the eastern side of peninsular India, for the assessment period 2018–2019. Highly correlated air pollutants and meteorological parameters with PM_2.5 concentration were evaluated and presented during the period under study. It was found that the CatBoost regression model outperformed all other employed regression models in predicting PM_2.5 concentration with an R² score (coefficient of determination) of 0.81, median absolute error (MedAE) of 6.95 µg/m³, mean absolute percentage error (MAPE) of 0.29, root mean square error (RMSE) of 11.42 µg/m³ and mean absolute error (MAE) of 9.07 µg/m³. High PM_2.5 concentration prediction results in contrast to Indian standards were also presented. In depth seasonal assessments of PM_2.5 concentration were presented, to show variance in PM_2.5 concentration during dominant seasons.

Bayesian estimators and highest posterior density credible intervals are obtained for two popular inequality measures, viz. the Gini index and the Bonferroni index in the case of the Dagum distribution. The study considers informative and non-informative priors, i.e. the Mukherjee-Islam prior and the extension of Jeffrey’s prior, respectively, under the presumption of the Linear Exponential (LINEX) loss function. A Monte Carlo simulation study is carried out in order to obtain the relative efficiency of both the Gini and Bonferroni indices while taking into consideration different priors and loss functions. The estimated loss proves lower when using the Mukherjee-Islam prior in comparison to the extension of Jeffrey’s prior and the LINEX loss function outperforms the squared error loss function (SELF) in terms of the estimated loss. Highest posterior density credible intervals are also obtained for both these measures. The study used real-life data sets for illustration purposes.