Forecasting the temporal distribution of passenger traffic on road transport based on harmonic analysis methods

Abstract

The method for constructing a point forecast of passenger traffic using harmonic analysis tools is considered. The technology is based on a time series model which includes trend, seasonal and cyclical components. Particular attention is paid to the choice of the number of levels of the series used to build the model, identify and evaluate the significance of its parameters, study the quality of the constructed predictive model and the adequacy of its initial data. The process of constructing the forecast is considered on the example of the temporal distribution of passenger traffic on road transport according to daily data on the example of the Sverdlovsk region for November 2021 - October 2022. All necessary statistical procedures are applied to identify and evaluate the parameters of the model, check its adequacy and accuracy. Based on the results obtained, short-term forecasts and conclusions of the study are made.