论文部分内容阅读
根据水稻三化螟(Tryporyza incersulas Walker)测报因素的要求,每一世代发生期选择60～82个因子、发生量选择83～94个因子做为预报方程的变量。选出因子后,应用IBM—PC/XT计算机进行多元逐步回归分析,建立预报方程:y_x=a_0x_0+a_1x_1+…+a_nx_n(y_x为发生期或发生量、a_n为系数、x_n为预报变量)。同时将测报站历年的综合防治经验编成程序,输入电脑,以供防治时调出参考。通过两年的实践验证,预报完全符合率达57.14％;允许误差范围内符合率达92.90％。
According to the requirements of the reporter for Tryporyza incersulas Walker, 60 to 82 factors are selected in each generation, and 83 to 94 factors are selected as the variables for predicting the equation. After selecting the factors, multiple stepwise regression analysis was performed with IBM-PC / XT computer to establish the prediction equation: y_x = a_0x_0 + a_1x_1 + ... + a_nx_n (y_x is the occurrence period or occurrence, a_n is the coefficient, x_n is the forecasting variable). At the same time measuring station over the years of integrated prevention and control experience compiled into a program, enter the computer for reference when the prevention and treatment. After two years of practical verification, the full coincidence rate of prediction reaches 57.14%; the coincidence rate within the allowable error range reaches 92.90%.