Since small and medium ¡Vsize enterprises (SMEs) have played a fundamental role in the economic development of this country, as a intermediary, the financial assistance of Taiwan SMEG to SMEs, is the greatest contribution to the above, nevertheless, Taiwan SMEG has been immersing in a predicament of financial deficit due to the unbalanced budget for a long time. Except the Package Credit Guarantee has introduced the concept of total risk control, all the others like the Authorized Approach and the Normal Approach have not introduced yet. For its perennial operation, Taiwan SMEG must set up an appropriate system for the rate of guarantee fees to achieve self-contained and self-sufficient condition, also to obtain the best efficiency under its limited resources. This paper uses the market ¡Vbased risk neutral model developed by Kuo (2006) to estimate the probability of default of banking loan assets through the risk premium of each banking loan, further to stimulate the guarantee fees by using the actuarial valuation principles. The purpose of this model is to react the degree of credit risk on the stimulated guarantee fees , making the fees pricing mechanism reasonable and fair . The model also analyzed the cost of subrogation payment under default cases by different industry , making it a reference for banks and Taiwan SMEG. The empirical results show that: 1.The credit guarantee fees stimulated by this model are approximately fair comparative to the actual subrogation payments of default cases reimbursed by Taiwan SMEG. The result provide evidence that the model possess the power of fitness for estimating the default cost of subrogation payment. 2.The degree of credit risk can actually react on the guarantee fees through using this model. It is essential to set up different range for rate of guarantee fees according to the subrogation payment ratio of default cost caused by respective guaranteed industry. The result suggests that Taiwan SMEG should amplify the extreme energy of guarantee through reasonable and fair use of its limited resources. Using the data of Taiwan SMEG guaranteed cases, this paper also analysis the default cost between the cost of funds and earning profits for the banking loans. The banks in Taiwan earn less profits than before under the overbanking environment. The competitive bank¡¦s loan pricing strategy leads to extreme loss while the default cases occur. In the lights of risk management , banks have become a high risk and low return industry. It is essential for banks to emphasize the loan quality and pricing strategy when expanding their loan business. Banks must simultaneously evaluate the RAROC under a perfect risk management system, so that the monetary environment can be improved and banks can take advantage of it by earning reasonable profits.

A main goal of the diploma thesis is to perform a strategic analysis of a selected enterprise with the purpose of examining a current situation and trends in external and internal environment of the enterprise, and, based on the analysis, to suggest recommendations concerning a future strategy of the company.

The current method for house valuations in mortgage portfolio models corresponds to applying a residential property price index (RPPI) to the purchasing price (or last known valuation). This thesis introduces an alternative house valuation method, which combines the current one with the bank's customer data. This approach shows that the gap between the actual house value and the current estimated house value can to some extent be explained by customer attributes, especially for houses where the homeowner is a defaulted customer. The inclusion of customer attributes can either reduce false overestimation or predict whether or not the current valuation is an overestimation or underestimation. This particular property is of interest in credit risk, as false overestimations can have negative impacts on the mortgage portfolio. The statistical methods that were used in this thesis were the data mining techniques regression and clustering. / De modeller och tillvägagångssätt som i dagsläget används för husvärdering i bolåneportföljen bygger på husprisindexering och köpesskilling. Denna studie introducerar ett alternativt sätt att uppskattta husvärdet, genom att kombinera dagens metod med bankens egna kunddata. Det här tillvägagångssättet visar på att gapet mellan det faktiska och det uppskattade husvärdet kan i viss mån förklaras av kunddata, framförallt där husägaren är en fallerad kund. Inkluderandet av kunddata kan både minska dagens övervärdering samt predicera huruvida dagens uppskattning är en övervärdering eller undervärdering. För fallerade kunder gav den alternativa husvärderingen ett mer sanningsenligt uppskattat värde av försäljningspriset än den traditionella metoden. Denna egenskap är av intresse inom kreditrisk, då en falsk övervärdering kan ha negativa konsekvenser på bolåneportföljen, framförallt för fallerade kunder. De statistiska verktyg som användes i denna studie var diverse regressionsmetoder samt klusteranalys.