Abstract:

In order to make accurate prediction on pecan yield,this paper takes Lin 'an pecan as an example for empirical study and analyzes the influential factors such as the climate and the planting quantity through establishing the Lasso regression and gray prediction combination model.The results show that compared with the prediction effects of the three models,Lasso-BP,Lasso-RBF and Lasso-GRNN,the prediction accuracy of the Lasso-GM model is higher,with an average relative error of 6.99 percent,from which the law of the changes of pecans output can be mastered.Meanwhile,the prediction results also show that by 2024,the output of Lin 'an pecans will continue to increase in a stable way.

Key words: Lin'an area, pecans, lasso-grey model, yield prediction