统计预报衣作物病虫害,近几年来在专业测报站已广泛的应用.我省临沂、济宁、汶上、惠民、聊城、烟台、德州、商河等地、县病虫测报站,运用多年积累的资料,采用统计预报方法,预报粘虫、小麦锈病、玉米螟、粟灰螟、大豆造桥虫、小地老虎、黄地老虎、棉铃虫、棉尖象等病虫害发生量和发生期,通过验证,一般准确率较高,这是实现测报科技现代化的手段之一,并为专业病虫测报站提供统计预报的方法.本文选择几种运算简便,准确率又较高的统计预报方法介绍如下预报要素分级多元线性回归分析法多元线性回归分析法,在病虫测报中是一种常用的统计预报方法.可以考虑较多的预报因子,同时也可估计到各预报因子的相互作用,因此比一元线性回归即直线回归法预报的结果更能接近实际.多元回归选取预报因子也不如简单回归要求严格,只要有一定程度的相关即可,因为在多元回归中这个 Statistical forecast clothing crop pests and diseases, in recent years in a professional measuring station has been widely used. Linyi, Jining, Wenshang, Huimin, Liaocheng, Yantai, Texas, Shanghe and other places, counties and pests stations, the use of accumulated over the years , The incidence and occurrence of pests such as armyworm, wheat rust, corn borer, corn borer, soybean bridge parasite, small tiger, yellow tiger, cotton bollworm and cotton spike were predicted by statistical methods. Verification, the general accuracy rate is higher, which is one of the means to realize the modernization of the newspaper science and technology, and provides the statistical forecast method for the professional pest reporting station.This article chooses several kinds of statistical forecast methods which are simple, accurate and high Predict factor classification Multivariate linear regression analysis Multiple linear regression analysis is a commonly used method of statistical forecast in pest forecasting.It can consider more forecasting factors and estimate the interaction of each forecasting factor, The result of one-linear regression, that is, the linear regression forecast, is closer to reality.Multivariate regression is not as good as simple regression in selecting predictors as long as there is a certain degree of phase You can, because in the multiple regression