Kalman model for short - term forecasting of air pollution levels ” ( 2005 )

“A Bayesian kriged Kalman model for short-term forecasting of air pollution levels” (2005) Sujit K. Sahu and Kanti V. Mardia http://www.blackwell-synergy.com/links/doi/10.1111/j.14679876.2005.00480.x/pdf?cookieSet=1 Shor t-term forecasts of air pollution levels in big cities are now repor ted in newspapers and other media outlets. Studies indicate that even shor t-term exposure to high levels of an air pollutant called atmospheric particulate matter can lead to long-term health effects. Data are typically obser ved at fixed monitoring stations throughout a study region of interest at different time points. Statistical spatiotemporal models are appropriate for modelling these data. We consider shor t-term forecasting of these spatiotemporal processes by using a Bayes-ian kriged Kalman filtering model. The spatial prediction surface of the model is built by using the well-known method of kriging for optimum spatial prediction and the temporal effects are analysed by using the models underlying the Kalman filtering method. The full Bayesian model is implemented by using Markov chain Monte Carlo techniques which enable us to obtain the optimal Bayesian forecasts in time and space. A new cross-validation method based on the Mahalanobis distance between the forecasts and obser ved data is also developed to assess the forecasting performance of the model implemented.