摘要
近年来涌现出大量基于DEA方法的银行效率这一热点问题的研究成果,但是由于DEA方法自身的局限性,诸如对效率值的估计偏低且离散程度较大,以及不能方便地检验结果的显著性等等问题,使其研究的结果受到影响;而Bootstrap技术是基于对经验数据及其相关估计的重复抽样来提高估计置信区间和临界值精度的统计技术,可有效地克服DEA方法结果可信度的这种内在依赖性。本文给出了基于Boot-strap技术的DEA方法来计量银行效率的手段,提高了效率分析结果的可信度,并对我国四大国有商业银行的效率进行了对比性实证分析。
DEA is a useful analytical way of testingbanking efficiency, but its application is constraint becausethe calculation of hte DEA efficiency score for one DMU involves all the other DMUs in the observation set. Moreover, Bootstrap is useful technology to overcome theshortcoming of DEA way and can improve the credit leveloftest, So we construct a immovation method to analysis thebanking efficiency with DEA and test with Bootstrap. Finalis Chinese four state-owned bank demonstration.
出处
《上海管理科学》
2005年第5期47-48,共2页
Shanghai Management Science
