Exploring the dynamics of arrivals and prices volatility in onion (Allium cepa) using advanced time series techniques

Abstract

Modeling the arrivals and prices of agricultural commodities is an essential requirement for farmers, consumers, and governmental organizations to make informed decisions. This is particularly important for perishable commodities such as vegetables, where spoilage can lead to significant losses for farmers and have a ripple effect on supply and demand dynamics. Volatility in the arrivals and prices of vegetables like onion is a serious issue affecting the common person in different ways. The study attempts to employ different time series models like the autoregressive integrated moving average (ARIMA), Artificial neural network (ANN), hybrid, and ensemble empirical mode decomposition (EEMD) techniques to analyze the pattern and trend of onions in Chandigarh and Delhi markets. From the results of the study, the amount of volatility in the data was found to range from medium to high among the markets. Decomposition techniques such as EEMD-ARIMA and EEMD-ANN performed better for the study data with the least mean absolute percentage error (MAPE) values, such as 17.74 and 6.78% for arrivals and 9.76 and 10.24% for prices at Chandigarh and Delhi markets, respectively. The EEMD techniques exceled in handling the non-linearity and non-stationarity by decomposing the data into different intrinsic modes and a residual, providing a better understanding of the fluctuation levels of data.