Abstract: The Bootstrap method was introduced into the DEA model. Under the conditions of constant return to scale and variable return to scale， the initial DEA efficiency values of the China 67 commercial banks were rectified before and after financial crisis. The efficiency values of the state owned commercial banks， joint-equity commercial banks and city commercial banks were compared. The Bootstrap truncated regression method was used to test the factors affecting bank efficiency. The empirical results showed that the DEA efficiency values adjusted by Bootstrap are significantly lower than the initial DEA values. Compared with the state owned commercial banks， the efficiency levels of joint-stock commercial banks and city commercial banks are greatly affected by the financial crisis. The bank technical efficiency can be improved by properly expanding bank size， increasing profitability and innovation ability， while excessive revenue expenses and excessively dispersed ownership structure will reduce the technical efficiency of banks.

Key words: bank efficiency, Bootstrap, DEA, truncated regression, technical efficiency