1
2
3
5Senior Scientist,
6Scientist,
*Corresponding author email id: nansnew@gmail.com
In this study, to model and forecast the onion and potato price in Lasalgaon and Agra market respectively during the period March 2009 to March 2019, Auto-Regressive Integrated Moving Average (ARIMA) and Generalized Auto- Regressive Conditional Heteroskedasticity (GARCH) have been used. Augmented Dickey-Fuller (ADF) and Phillips- Perron (PP) test were used for checking the presence of unit root in the time series. Akaike Information Criterion (AIC) values were used for identifying the suitable ARIMA model. The residuals of the fitted ARIMA models revealed the presence of autocorrelation and ARCH effects, so GARCH model is also used for forecasting. Both the models were compared with respect to forecast accuracy measures. In both the cases, it is found that GARCH model performed better than ARIMA model.
Forecasting, Agricultural market, ARIMA model, GARCH model