Spillover Effects of Covid-19 Induced Lockdown on Onion Prices in India

Abstract

In a normal situation onion prices vary in a highly unprecedented way in India. So, it is worth noticing the effect of an uncertain situation on onion prices. In this article an efficient Artificial Intelligence (AI) tool, i.e., Support Vector Regression (SVR) has been used to predict the price fluctuation of onion over the lockdown period, unlock condition and the period including the pre-pandemic situation. Results obtained are compared with prediction of traditional Multiple Linear Regression (MLR) model. Several metrices such as , Root Mean Squared Error (RMSE), Mean Absolute Deviation (MAD), and Relative Mean Absolute Percentage Error (RMAPE) have been used for this purpose. The result of Machine Learning (ML) algorithm indicates that in the nationwide lockdown condition pandemic indicator variables are having more than 70% influence on the onion price variability. The effect is reduced to near about 60% in unlock condition and if considering the whole year data this effect is near about 45%. The results also indicate that ML algorithm is more efficient to capture the variability than the traditional model.