Analyzing the Effect of Dual Long Memory Process in Forecasting Agricultural Prices in Different Markets of India

Abstract

The potential presence of long memory (LM) properties in mean and volatility of the spot price of wheat and mustard in different markets of India has been investigated. The findings revealed the evidence of long range dependence in price series as well as in the volatility. Accordingly, Autoregressive fractionally integrated moving average (ARFIMA) with error following Fractionally integrated generalized autoregressive conditional heteroscedastic (FIGARCH) model has been applied for forecasting the price of commodities in different markets of India. To this end, evaluation of forecasting is carried out with root mean squares error (RMSE), mean absolute error (MAE) and relative mean absolute prediction error (RMAPE). The residuals of the fitted models were used for diagnostic checking.