Modelling and Forecasting Wheat Production in Punjab State of India using Hierarchical Time Series Models

Abstract

Due to the importance of wheat crop and Punjab being the leading wheat producer, this paper considers hierarchical time series data on wheat production in Punjab. The Punjab state wheat production data is organized in a hierarchy based on geographical regions. Topdown, bottom-up, middle-out, and optimal-combination approaches were used along with single series forecasting. An analysis of forecast performance shows that bottom-up approach outperforms other methods in terms of RMSE and MAE in out of sample forecast horizon. Finally, the state of Punjab has forecasted wheat production from 2019 to 2023 using a bottom-up approach.