The early warning system of banking industry-the application of market and accounting information

碩士 === 東吳大學 === 經濟學系 === 91 === Abstract Due to the Government liberalized the establishment of bank; the banks of Taiwan had become oversupply. To maintain operation, the banks sacrifice the interest spread, moreover the recession and finance crisis worsen the bank’s performance. As we k...

Full description

Bibliographic Details
Main Authors: Wen-Chung Liu, 劉文仲
Other Authors: Chia-Hsiang Guo
Format: Others
Language:zh-TW
Published: 2002
Online Access:http://ndltd.ncl.edu.tw/handle/05730713744776403296
Description
Summary:碩士 === 東吳大學 === 經濟學系 === 91 === Abstract Due to the Government liberalized the establishment of bank; the banks of Taiwan had become oversupply. To maintain operation, the banks sacrifice the interest spread, moreover the recession and finance crisis worsen the bank’s performance. As we know, the source of bank’s fund is based on the saving of the depositors. So the depositors and government concern about the management performance of the bank, therefore the banks are highly regulated by the government. Therefore, our financial monitoring institution should set up a good system of financial supervision and precaution system to avoid the banks go to bankruptcy. The paper uses the variables, which could reflect the bank’s market asset value and combine the accounting variables to setup the bank performance evaluation and early warning system. Compare its result with only use the accounting variables and discuss the difference between two kinds of Logit model. Some conclusions are drawn from my empirical study: 1. After factor analysis, the market variables are remained and accumulated enough power to fit the bank’s performance evaluation. 2. The Logit model constructed by factor analysis shows that the deviation of the return rate of market asset value is reserved at four prediction models. So the market variables are essential to setup the prediction model. 3. For the power of prediction, the result of Logit model constructed by factor analysis shows that use the lately market variables works better than general model. 4. Seemly, for the power of prediction, the result of the stepwise logistic regression model, which uses the market variables, is less effective than general model.