Comparison of Time Series Forecasting Models in Garlic's Wholesale Price

Garlic is an important seasoning vegetable that can not be excluded from Korean diet. Predicting its supply and demand situations and price is very important in terms of producer's income and consumer price stability. This study estimated the error correction model (ECM) and the Bayesian VAR model using time series price data of garlic. Also this study assessed the predictive power of the estimated model by performing the out-of-sample forecasts. All price data used in the analysis were identified as non-stationary time series data. There was a cointegration relationship between wholesale prices of whole bulbs of garlic and peeled garlic, so the error correction model and the Bayesian VAR model were estimated. Estimation results showed that predictive power of the models was pretty good and the error correction model had better predictive power than the Bayesian VAR model. The estimated garlic pricing models in this study are expected to contribute not only to the current price prediction model based on quantity forecasting but also to the efficiency of the model operation process.