This research aims to demonstrate how discriminant function analysis may be used to create a rice production forecasting model for Jaunpur (India). Discriminant function analysis is a method of creating a linear/quadratic function that best discriminates different populations and so provides a qualitative evaluation of the likely yield. Time series data from 18 years (2000-2019) have been divided into three categories: congenial, normal, and adverse, based on yield distribution. Taking these three groups as three populations, discriminant function analysis has been carried out. The regressors in the model were discriminant scores obtained from this. The use of weekly weather data has been proposed in a variety of ways. The models were used to forecast yields for the three years following 2015-16: 2015-17: 2017-18. (which were not included in model development). About two months before harvest, the method offered a reliable yield prediction.