Time series analysis of castor crop for price forecasting in Gujarat: A comprehensive study

Abstract

The overall objective of the present paper is demonstrating the utility of price forecasting of farm prices and validating the same for castor crops in Gujarat state for the year 2022 using the time series data from 2007 to 2021. While for price data of castor was collected from AGMARKNET (www.agmarknet.gov.in). Looking to the seasonal indices, the lowest and highest seasonal indices of castor price were happened in May and August respectively for all the castor markets except Patan market. The results showed that the lower instability for price under castor was observed in Mehsana (1.045), Rajkot (0.999), Gandhinagar (1.158), Banaskantha (1.074), Patan (1.903) and Sabarkantha (1.089) districts. Majority of the districts showed the low level of instability for price. There may be chance of volatility persist in these markets yet it should be subjected to formal ARIMA effect test to confirm the presence of volatility. The seasonal component was estimated by fitting the cubic trend in Mehsana, Rajkot, Gandhinagar, Banaskantha and Sabarkantha markets. However in Patan market the trend observed in compound, growth, exponential and logistic. The results were obtained from the application of univariate ARIMA techniques to produce price forecasts for castor crop and precision of the forecasts were evaluated using the standard criteria of lower value of RMSE, MAPE, MAE MSE with higher value of Adj. R2. On the basis of these criteria find out best model of ARIMA for castor price forecasted. Among the selected six markets, Mehsana ARIMA (0,1,0), Rajkot ARIMA (0,1,2), Gandhinagar ARIMA (0,1,0), Banaskantha ARIMA (0,1,0), Patan ARIMA (1,0,1) and Sabarkantha ARIMA (1,1,1) model were found to be best fitted for the forecasting the price of castor in Gujarat.