Vector autoregressive time series model in predicting food prices in Manado city
Open Access
- 1 January 2023
- conference paper
- Published by AIP Publishing in AIP Conference Proceedings
- Vol. 2694 (1)
Abstract
This study discusses the Vector AutoRegressive (VAR) model for multivariate time series. The VAR model is the development of a univariate model into a multivariate time series model. The purpose of the research is to model the prices of food ingredients in the city of Manado, namely rice, cooking oil, chicken eggs, chicken meat and cayenne pepper. This is important because in the global era with the rapid development of science and technology, the influence of the price of one staple food will affect the price of other staples so that the autoregression vector model can be applied. The data analyzed is monthly data from January 2015 to December 2020. For parameter estimation and statistical analysis, a computer program with algorithms and coding of parameter estimation programs is used. By using the Box-Jenkins three-stage methodology, namely identification, parameter estimation and model diagnostics, we obtain a VAR model and price movements that are suitable for the data. Prices of rice and cooking oil showed a more stable pattern, while the price of cayenne pepper was more volatile. Furthermore, by using the model, predictions of prices and price movements for the future period can be obtained.Keywords
This publication has 5 references indexed in Scilit:
- Vector Autoregressive Models with Multivariate Time Series: Spillovers of International Tourist Arrivals on a Local Economy2018
- Consistency and asymptotic normality of least squares estimators in generalized STAR modelsStatistica Neerlandica, 2008
- Modeling Financial Time Series with S-Plus®2003
- Analysis of Financial Time Series2002
- Time Series Analysis1994