Computational modeling for forecasting of operational holdings and area in India

Abstract

Various computational modeling techniques such as ARIMA (Box-Jenkins), State-Space, Structural and other time series depending upon the properties of the given time series from agriculture are now being effectively utilized. Modelling and related forecasting for the time series data of Agriculture Census (2015-16) of the country were performed using Auto-Regressive Integrated Moving Average (ARIMA) model. ARIMA (0, 1, 0) and ARIMA (0, 2, 0) were found suitable for different size groups of operational holdings and their operational area based on the least value of Akaike Information Criterion (AIC). The total number of operational holdings in the country are predicted to be increased from 146 million in 2015-16 to 154 million in 2020-21 and 161 million by 2015-26. Besides, there would be decrease in the operational area from 157.14 million ha in 2015-16 to 155.91 million ha and 154.74 million ha by 2020-21 and 2025-26, respectively. As such, the average size of operational holding is expected to decline to the level of 1.01 ha in 2020-21 and further to 0.96 ha by 2025-26 in comparision to 1.08 ha of 2015-16. The small and marginal farmers holding less than 2 hectares are expected to be 87.34% and 88.35 % with a share of 50.59% and 53.89% in operational area by 2020-21 and 2025-26, respectively. These information’s would thus, play a crucial role in development planning, socio-economic policy formulation and the establishment of national priorities.