Author(s): Ajay Sharma and Joginder Kumar

Abstract: Potato (Solanum tuberosum L.) price volatility affects consumers, producers, and policymakers to support the needs of stakeholders. Therefore, the present study is planned to fit the statistical models for the analysis of volatility in potato prices of selected markets of Northern India using advanced time-series models such as Standard Generalized Autoregressive Conditional Heteroskedasticity (SGARCH), Glosten, Jagannathan, and Runkle Generalized Autoregressive Conditional Heteroskedasticity (GJR-GARCH) and Artificial Neural Network (ANN). For this, the time series data of monthly wholesale potato prices from January 2010 to December 2023 have been used for fitting and analysing the statistical models. The accuracy of the fitted models has been evaluated using various statistical measures, including the Akaike Information Criteria (AIC), the Bayesian Information Criteria (BIC), the Root Mean Square Error and the Mean Absolute Percentage Error (MAPE). The findings of the present study revealed that ARMA (1,1)-GJR-GARCH (1,1), ARMA (1,1)-GJR-GARCH (1,1), ARMA (1,1) -SGARCH (1,1), ANN (12-6-1), ANN (12-8-1) and ANN (12-6-1) models have been found better among various fitted models for estimating the potato price volatility in selected monthly wholesale markets of Northern India. However, the effectiveness of the fitted models has also been evaluated using Relative Deviation (RD %), and found that the GJR-GARCH model has low RD (%) values as compared to other fitted models. This modelling approach can provide valuable insights to aid decision-making for farmers, traders, policymakers, and researchers involved in agricultural economics and market planning.

Pages: 163-171 | Views: 135 | Downloads: 4

Download Full Article: Click Here