LINEAR MIXED MODELS FOR SUGARCANE YffiLD ESTIMATION INHARYANA (INDIA)

摘要

The aim of this paper is to show the usefulness of mixed modelling approach for crop yield prediction. The study deals in obtaining the pre-harvest crop yield forecasts of major sugarcane growing districts in Haryana. The common approach to crop yieldforecast is linear regression with constant coefficients over time. This may be restrictive and of limited prediction power since it does not account for year-to-year dependence in the yield variable. However, a mixed model procedure provides a flexibleway to fit multi-level models for crop yield prediction. The linear mixed effects models with random time effects at district, zone and state level were fitted for sugarcane yield estimation in Haryana. The percent deviations of district-level sugarcaneyield forecasts from the real time yield(s) show a preference for using linear mixed models.