客戶信用評等個案分析

碩士 === 國立政治大學 === 經營管理碩士學程(EMBA) === 101 === The major purpose of this study is based on the operation needs of the case company, to design an early warning model to measure its customers credit risk -「Credit Score Card」, wish to use this model to predict the probability of customers’ financial cris...

Full description

Bibliographic Details
Main Authors: Lee, Anne, 李美玉
Other Authors: 姜堯民
Format: Others
Language:zh-TW
Online Access:http://ndltd.ncl.edu.tw/handle/65878309031207085454
Description
Summary:碩士 === 國立政治大學 === 經營管理碩士學程(EMBA) === 101 === The major purpose of this study is based on the operation needs of the case company, to design an early warning model to measure its customers credit risk -「Credit Score Card」, wish to use this model to predict the probability of customers’ financial crises, to achieve the early warning purpose. The direction of the study is the use of financial and non-financial variables to build up the 「Credit Score Card」, and using the customers’ data of case company to test its effectiveness. The empirical results are as follows: 1. This study suggests that 「Credit Score Card」 has the ability to predict the corporate financial crises. Each indicator's impact on the financial crisis early warning are negatively correlated, ie, the better performance of the indicators, the lower the probability of the financial crisis occur. 2. Risk cannot be completely eliminated, but can be controlled, through the establishment of measurable indicators to assess the risk, can help to reduce the uncertainty and possible future loss. 3. When an organization is highly decentralized and regionalized, it must have unified credit risk management policies and measures to facilitate the regions to follow. 4. Risk metrics or weight should depend on the sales market, business model and industry characteristic, for optimal correction. 5. If it is difficult to obtain data for indicator measurement, it will affect the effectiveness of indicator, therefore need to be cautioned when screening indicator measurement.