Forecasting prices of coffee seeds using Vector Autoregressive Time Series Model

Abstract

Forecasts of agricultural prices are useful to the farmers, policymakers and agribusiness industries. In this globalized world, management of food security in the developing countries like India where agriculture is dominated needs efficient and reliable price forecasting models. In the present study, Vector Autoregression (VAR) has been applied for modeling and forecasting of monthly wholesale price of clean coffee seeds in different coffee consuming centers, viz. Bengaluru, Chennai and Hyderabad. Augmented Dickey-Fuller (ADF) test has been used for testing the stationarity of the time series. The appropriate VAR model is selected based on minimum Akaike Information Criterion (AIC). The VAR model obtained is compared with the Auto Regressive Integrated Moving Average (ARIMA) models with respect to forecast accuracy measures. The residuals of the fitted models were diagnosed for possible presence of autocorrelation and Autoregressive Conditional Heteroscedasticity (ARCH) effects.