Forecasting of pre-harvest crop yield using discriminant function analysis of meteorological parameters

Abstract

In the present paper, an application of discriminant function analysis of meteorological parameters for developing suitable statistical models to forecast wheat yield in Faizabad district of Eastern Utter Pradesh has been demonstrated. Time series data on wheat yield for 20 years (1990-91 to 2009-10) have been divided into three groups, viz. congenial, normal, and adverse based on de-trended yield distribution. Considering these groups as three populations, discriminant function analysis using weekly data of crop season on five meteorological parameters has been carried out. The discriminant scores obtained from this have been used as regressor variables along with time trend in development of statistical models. In all six procedures using weekly weather data have been proposed. The models developed have been used to forecast the wheat yield for the year 2008-09 and 2009-10 which were not included in the development of the models. It has been found that most of the models provide reliable forecast of the wheat yield about two months before the harvest. However, the model-5 has been found to be the most suitable among all the models developed.