基于灰色关联分析的熵权法在企业绩效评价中的应用

Application of entropy weight method based on grey cognate analysis in enterprise performance evaluation

针对综合评价方法在数据标准化处理和权重设置上存在的问题,提出基于灰色关联分析的熵权法,借助灰色关联度分析理论,对熵权法进行修改和完善,即采用计算指标灰色关联系数的方式对原始数据进行标准化处理,并用该评价模型对上市房地产开发企业进行绩效评价。通过实例分析发现,基于灰色关联分析的熵权法能一步到位解决指标的一致化处理和指标的无量纲化问题,同时最大程度地保留了原始数据所携带的信息,确保了熵权法的准确性。该评价模型发挥了灰色关联度分析法和熵权法的优势,解决了评价过程中评价主观、指标量纲不同、指标逆向等问题,使评价结果更加科学、合理,并且模型算法具有良好的鲁棒性,评价对象的增减不影响企业绩效的相对排名。

The entropy weight method based on grey relational analysis is proposed to solve the existent issues of comprehensive evaluation method on data normalization treatment and weight setting. The entro- py weight method is modified and improved by means of the grey relational degree analysis theory. In this method, the original data was standardized, by means of calculating the gray correlation coefficient of the original data. The evaluation model is used to evaluate the performance of listed real estate development enterprises. These findings are discovered by example on grey relational analysis, could solve the consistency analysis that the entropy weight method, based updisposal and non-dimensional problem of the index in one step. Simultaneously, the method reserves the initial data message to the maximum extent and insures the accuracy of the entropy weight method. This assessment model gives full play to the grey relational analysis and entropy weight method＇ s advantages and settles issues like subjective appraisal, different index measurements and reverse index during the estimation process, which makes the appraisal resuit more scientific and reasonable. Furthermore, the model calculation method has excellent robustness and the assessed objects increase or decrease will not affect the relative rankings of the enterprise performance.