Potato Yield Classification Using Weather Variables: a Discriminant Analysis Approach

The study aims to develop a multivariate statistical model to classify potato yield based on weather variables. The potato yields of India from years 1950 to 2021 were classified into two and three categorical groups based on yield adjusted for trend. Annual and seasonal weather variables such as minimum temperature, maximum temperature and rainfall were used as independent variables to predict the classification of potato yield. Linear discriminant analysis was employed to study the classification of potato yield using weather variables. Results revealed that weather variables like monsoon rainfall and post-monsoon rainfall positively influenced the discrimination, whilst variables like annual minimum temperature and annual rainfall negatively influenced the discrimination of potato yield. Weather variables like post-monsoon minimum temperature and post-monsoon maximum temperature did not influence the discrimination of potato yield. Results revealed that discrimination of potato yield using three groups performed better than two groups with 90.3% accuracy. Discrimination of yield using weather variables gives a better understanding of the influence of weather on potato yield quantitatively throughout the year. It helps to predict the group membership of the yield.