Application of Projection Pursuit Regression in the Forecast of the Percentage of Damaged Seeds of Leguminivora Glycinivorella

Abstract

In the paper, we build the forecast model of the percentage of damaged seeds of leguminivora glycinivorella by the way of projection pursuit regression, because the percentage of damaged seeds of leguminivora glycinivorella is affected by several factors, such as the cardinal number that leguminivora glycinivorella live through the winter and weather conditions and so on, and it is a nonlinear and abnormal system. The historical fitting rate and the forecast fitting rate are both satisfactory. So projection pursuit regression can be a new method of the insect pest forecasting.