Econometric analysis of price fluctuations in onion crop using SARIMA model
Abstract
The focus of the present study is to model and forecast the conditional monthly prices of onionin Pakistan by applying SARIMA(p,d,q)(P,D,Q)s model.Secondary data of monthly average prices were collected from Pakistan Bureau of Statistics (PBS). The data were first checked for stationarity though ADF test and found non-stationary hence, first difference was taken to make the data stationary.The differenced prices were then modeled by applying Box-Jenkins methodology. Different specifications of seasonal ARIMA i.e., SARIMAmodels were used and the most suited model was selected based on the AIC, BIC, and the white noise property of the residualsresulted from all the candidates models.SARIMA (1,1,1)(1,0,1)12 was selected as the best model and used to forecast the future prices. On an overall basis, the forecasted prices showed an increasing trend with the peaked prices in September-October which is due to the demand and supply gap of onion in the markets. It is concluded that such verities of onion should be grown which are available in market during the peak price months so that the price hike can be controlled. The present study can be further improved by applying structural change model.
Key words: Econometric modelling, Forecasting, Onion, Seasonal ARIMA.
