Forecasting air passengers at São Paulo International Airport using a mixture of local experts model

An integrated mixture of local experts model is employed to forecast air passengers at São Paulo International Airport. Such approach is normally used in rapidly changing situations, i.e., when the time series presents turning points or any kind of structural change and allows the development of forecasting models that takes into account the heterogeneity of the mapping structure into different regions of the input space. The model is validated using out-of-sample data, and the accuracy of the generated predictions proves to be satisfactory. An assessment of uncertainty in the predictions is made, as well as long lead-time forecasts employing the built model, considering different scenarios.

► A forecasting model of monthly passenger numbers at the São Paulo International Airport Brazil.
► The accuracy of the predictions of the short-term turning points proved to be satisfactory.
► Uncertainty in predictions are based on various GDP growth scenarios.