Deep particulate matter forecasting model using correntropy-induced loss

Forecasting the particulate matter (PM) concentration in South Korea has become urgently necessary owing to its strong negative impact on human life. In most statistical or machine learning methods, independent and identically distributed data, for example, a Gaussian distribution, are assumed; however, time series such as air pollution weather data do not meet this assumption. study, maximum correntropy criterion regression (MCCR) loss is used an analysis of characteristics data. Rigorous seasonality adjustment was performed because their complex patterns heavy-tailed distribution even after deseasonalization. The MCCR applied multiple models including conventional state-of-the-art models. results show that more appropriate than mean squared error forecasting extreme values.