摘要
通过测算贷款、存款等投入要素对净利息收入的贡献,评价商业银行的投入产出效率,对银行的资本运营和监管机构的银行资本监管具有重要意义.原始投入变量过多和变量之间的高度相关都会对评价模型的估计和检验产生影响.创新和特色在于:一是通过提取互不相关的2个主成分,反映6个原始投入变量95%以上的信息.建立基于主成分的SFA模型,克服变量过多和变量高度相关对模型参数估计和检验的影响,解决原始投入变量高度相关导致的系数检验不显著和符号不正确问题.二是利用主成分回归,将主成分与投入变量的关系表达式代入基于主成分的SFA模型,进而确定投入变量的权重系数,建立银行的投入产出模型,反映6个投入变量对净利息收入的影响规律.实证研究结果表明:一是利用主成分建立的SFA模型系数检验显著,技术效率随时间增加.二是利息支出、贷款余额、总资产、存款总额、固定资产和员工人数产出弹性分别为0.287,0.272,0.254,0.086,0.072和0.053.因此影响银行净利息收入的主要因素为利息支出、贷款余额、总资产.存款总额、固定资产和员工人数对净利息收入的影响较小.三是18家商业银行的规模系数为1.025,银行的净利息收入表现出规模经济特征.
It is of significant importance to evaluate the input-output efficiency of the com-mercial banks and assess the effects of loans, deposits and other input factors on net interest income. The results of model estimation and tests for a SFA model are likely to be affected by the number of input variables and the high correlation among variables. This study is innovative as it has two important features. Firstly, two unrelated principal components was extracted which reflected 95 percentage information of six original input variables. Based on the principal components, a SFA model, which can overcome the multilinearity problem on coefficient tests, was established to assess the bank efficiency. Secondly, an input-output model was setup based on the principal component regression. The weighted coefficient of input variable was determined by substituting the function of principal component about in- put variables into principal component-SFA model, which reflected the effect of the 6 input variables on the net interest income of banks. Empirical findings showed that the main fac- tors affecting the bank's net interest incomes were interest expenses, the loan balance, total assets. The effect on the net interest income of total deposits, fixed assets and the number of employees were non-vital. The output elasticity of interest payments, loans, total assets, total deposits, fixed assets and the number of employees was 0.287, 0.272, 0.254, 0.086, 0.072 and 0.053 respectively. The scale coefficient 18 sample commercials banks were 1. 025, which showed that the bank's net interest income demonstrated increasing return of scale.
出处
《数学的实践与认识》
CSCD
北大核心
2013年第8期62-75,共14页
Mathematics in Practice and Theory
基金
国家自然科学基金(71273044
71171031
71201018)
关键词
商业银行效率
投入产出效率
主成分回归
SFA模型
input-output efficiency of commercial bank
efficiency evaluation
principle-regression
the SFA model