Forecasting of Black Pepper Price in Karnataka State: An Application of ARIMA and ARCH Models

Abstract

The study was conducted to forecast the price of black pepper in one of the major markets of Karnataka state as the state ranks first position in production of pepper in India. The Gonikoppal market in Kodagu district was selected purposively on the basis of highest area and production in the state. The monthly prices of black pepper in Gonikoppal market were collected from the Karnataka State Agricultural Marketing Board, Bangalore, Karnataka state for the year 2008-09 to 2017-18. The time-series models such as ARIMA and ARCH models were applied to price data using software's such as SPSS, Gretl and EViews. The Augmented Dickey-Fuller test and Heteroscedasticity Lagrange's Multiplier test were used to test the stationarity and volatility of the time-series respectively. The best forecasted model was determined based on the lowest values of Akaike's Information Criterion (AIC) and Schwartz Bayesian Information Criterion (SBIC). However, the predictability power, performance and quality of the model was measured based on the lowest error value of the Root Mean Square Error (RMSE) and Mean Absolute Prediction Error (MAPE). Among the tested models the prediction accuracy of the ARIMA model was higher than ARCH family models. On the basis of the results, the ARIMA (0,1,1) provide a good fit for forecasting the price of black pepper.

Using Econometric Models to Manage the Price Risk of Cocoa Beans: A Case from India

“…Hence, it is advised to use information criteria statistics of the estimates to decide on the number of lags or parameters for the model. There are three popular information criteria recommended by many econometricians for optimal lag selection; they are, Akaike's information criteria (AIC), Schwarz's Bayesian information criteria (SBIC), and the Hannan-Quinn criterion (HQIC) (Agung 2009;Fabozzi et al 2014;Mallikarjuna et al 2019;Mills and Patterson 2009;Shmueli and Litchtendahl 2016). The ACF and PACF correlograms of the residuals of the estimates are used for the diagnostic checking (Gujarati et al 2009).…”

Application of Machine Learning Techniques Models for Forecasting of Redgram Prices of Andhra Pradesh, India

“…In comparison to the RF (M5) model, the SVR (M4) ANN (M3), GARCH (M2) and SARIMA (M1) models were significantly different. It means that RF model outperformed remaining models significantly [27,28]. The ability of RF model to outperform SVR, ANN, GARCH and SARIMA in the training set was due to its superior capacity of the model and the non-linear nature of the time series data of redgram prices data under consideration.…”