Modelling of Arima for the Production of Potato in India Using the Data from 1950 To 2022

Agriculture is the backbone of India which depend upon agriculture as a major occupation, farmers having accurate agricultural production forecast are very vital. Indirectly forecasting agricultural production which benefits policy makers and businessmen. Arima auto regressive integrated moving average method or a model, it is a powerful tool for data mining time series forecasting. Auto regressive captures the influence of historical values to the present value, integrated which helps in differencing and achieving stationary in data which is required for forecasting in ARIMA and moving average which refers to random errors, smoothening the data which in turn helps in improving forecasting accuracy. Arima can analyze various types of data namely the data which follows trends, seasonality, or seasonal patterns and non-seasonal. It helps in housework the data or finding the random mistakes which makes the data suitable for forecasting. In this paper by using historical production data of potato in India we aim to forecast underlaying trends and seasonality, and try to generate accurate forecasts for the future production of potato, and evaluate the model’s performance. The result obtained using ARIMA can be very vital and empower stakeholders to make informed decisions, also helping the farmers adjusting their planting strategies, policy makers preparing for production fluctuations, and the businessman optimizing or restructuring their supply chains based on the expected potato production.