DOI: 10.32725/978-80-7394-976-1.35

Consumer prices forecasting based on ARIMA models

Ondřej Šimpach

Prague University of Economics and Business, Faculty of Informatics and Statistics, Department of Statistics and Probability, W. Churchill Sq. 1938/4, 130 67 Prague 3, Czech Republic, ondrej.simpach@vse.cz.

High consumer prices and high inflation afflicted many countries in the world recently. The cause is the Covid epidemic and Russian aggression in Ukraine, which primarily led to an increase in energy prices. In the Czech Republic, the inflation reached 17.5% in June 2022. The rapid increase of consumer prices has consequences. The number of poor people that depend on social benefits is growing. Government prepares measures and subsidies and for low income households. For policymaking and planning, a projection of the consumer prices development in the future is needed. Therefore, the aim of the paper is to project the prices of basic food: bread, butter, milk, poultry, potatoes, sugar, and eggs. The data were taken from Czech Statistical Office with monthly frequency for the period of 2010–2022 in order to have long time series. We applied Box-Jenkinson methodology – SARIMA and ARIMA models and modelled the development of the time series and forecast to the future 12 months. For bread prices forecast, an ARIMA model is preferred. In case of butter, both models forecast slight decrease of price. and we cannot clearly conclude which model is better. In case of milk projection, SARIMA is in contrast with ARIMA, but both models could be realistic as the price of milk is very volatile throughout the whole period. For price of chicken and potatoes, both forecasts are in contrary, but ARIMA is more probable. SARIMA model projects more realistic development of sugar price. In case of eggs price, both models project stabilization, so it cannot be concluded which model is preferred.However, econometric correctness and the best fit of the model is not a guarantee that projected values would meet the expectations based on the theory and practical experiences.

Klíčová slova: ARIMA models, consumer prices, forecasting, inflation

stránky: 236-243