论文部分内容阅读
加权综合法是对两种或两种以上预测结果进行加权合并的方法,使原模型得以修正,提高其预测精确度。本文以粉尘接触水平与尘肺发病关系,用寿命表算得发病概率后建立logistic回归及灰色系统模型预测粉尘允许浓度,并在此基础上应用加权综合法预测粉尘允许浓度,其结果为:加权综合法所测允许浓度为:10.5mg/m ̄3,logistic回归为10.2mg/m ̄3,灰色模型为13mg/m ̄3;经拟合优度比较,加权综合法(R ̄2=0.9933,ε ̄2=0.0185)较logistic回归(R ̄2=0.9930,ε ̄2=0.0193)和灰色模型(R ̄2=0.8638,ε ̄2=0.3735)更为精确。
The weighted synthesis method is a method of weighted combination of two or more forecast results, so that the original model can be amended to improve the prediction accuracy. In this paper, the relationship between the level of dust exposure and the incidence of pneumoconiosis, life expectancy calculated after the probability of occurrence of the establishment of logistic regression and gray system model to predict the allowable concentration of dust, and on this basis, the use of weighted synthetic method to predict dust allowable concentration, the results are: The measured allowable concentration: 10.5mg / m ~ 3, logistic regression was 10.2mg / m ~ 3, the gray model was 13mg / m ~ 3; compared to the goodness of fit, weighted synthesis (R ~ 2 = 0 .9933, ε ~ 2 = 0.0185) than the logistic regression (R¯2 = 0.9930, ε¯2 = 0.0193) and the gray model (R¯2 = 0.8638, ε¯2 = 0.3735 ) Is more accurate.