全寿命费用的DEA优化逐步回归分析法及其应用

在全寿命费用估算建模领域，逐步回归能对多因素进行筛选，克服了建模因素之间的多重相关性对最小二乘估计的干扰。逐步回归分析法本质上是对数据平均趋势的估算，无法避免“由失真数据得到无效模型”的根本问题。为此，提出用数据包络分析（DEA）优化逐步回归的建模方法，用DEA方法对数据进行评价，剔除无效数据，将有效的数据用来逐步回归建模。该方法能有效克服干扰数据对提取成分的影响，弥补逐步回归方法的不足。通过实例计算并与逐步回归、多元线性回归比较分析，DEA优化逐步回归建模精度为2．64％，高于逐步回归的3．39％和多元线性回归的3．49％，具有实用价值。

Optimized Stepwise Regression Analysis with DEA and Its Application

Stepwise regression .filters independent variables while modeling, which overcome the multi-correlation between independent variables that disturb the least squares estimation in the life cycle cost modeling. The regression statistical modeling approach estimates the average trend of the data essentially, which cannot avoid the problem that obtain the wrong model according to the error data at all. Therefore, the optimized stepwise regression with Data Envelopment Analysis （DEA） was proposed to overcome the problem, which can evaluate the efficiency of the data and remove the ineffieiency data for stepwise regression modeling The new approach can overcome the effect of the disturbed data when.filtering principal components in the independent variables. Compared with the method of stepwise regression and multiple linear regression, the precision of the optimum stepwise regression with DEA is 2.64%, which is higher than stepwise regression with 3.39% and multiple linear regression with 3. 49%.