Forecasting of Air Passengers using Singular Spectrum Analysis

Abstract

Air transportation is the most appropriate option for extremely vast distances, such as those between cities, provinces, and countries. While unpredictability, high volatility, and seasonality sometimes result in complex behavior in air passenger time series, this research applies the Singular Spectrum Analysis technique for air passengers data and uses the linear recurrent type for forecasting. Trends, seasonality, cyclists, and noise can all be found and extracted using Singular Spectrum Analysis. Singular Spectrum Analysis has the potential to be a highly effective forecasting method.