None

On a basis of the development of credit risk models, this study aims to help managers of financial institutions understand the development of the models so as to develop their own model that will provide objective and reasonable references for banks to decide the lending rate. Furthermore, this study used "Utilization of Long-Term Funding" as the object and studied individual cases of approved loans. By using risk neutral evaluation method to study the difference between the lending rate of loans and the risk-free interest rate of public bonds, to extract implied probabilities of default and required credit risk premiums form actual market data on interest rates. These credit risk premiums of model were used to be compared with the actual markups of banks and the results are as follows: 1.Most values stated in credit risk premium are lower than the actual markups for banks usually consider the burden of other capital costs and the factor of liquidity premium when they set the rating for markup. 2.After a loan is approved, the assumed recovery rate upon application will adjust according to the market value of the collateral. When the recovery rate decreases, the expected loss rate on the loan will gradually increase. Moreover, the higher the assumed recovery rate, the larger the corrected expected loss rate after the loan is approved. 3.In recent years, the non-performing rate for banks in Taiwan has reached a record high. Even though banks face less credit risks when they make long-term loans in "Utilization of Long-Term Funding", the probability of default has increased in recent years, which has contributed to the increase of expected loss rate on the long-term loan. In sum, banks still face credits risks that should not be ignored when they manage long-term loans. Thus, it is necessary to improve loan review to enhance the quality of loans and to increase the efficiency of utilization of long-term fund.

This dissertation consists of three essays on credit risk models and their Bayesian estimation. In each essay, defaults or default correlation models are built under one of two main streams. In our first essay, sequential estimation on hidden asset value and model parameters estimation are implemented under the Black-Cox model. To capture short-term autocorrelation in the stock market, we assume that market noise follows a mean reverting process. For estimation, two Bayesian methods are applied in this essay: the particle filter algorithm for sequential estimation of asset value and the generalized Gibbs sampling method for model parameters estimation. The first simulation study shows that sequential hidden asset value estimation using option price and equity price is more efficient and accurate than estimation using only equity price. The second simulation study shows that by applying the generalized Gibbs sampling method, model parameters can be successfully estimated under the model setting that there is no closed-form solution. In an empirical analysis using eight companies, half of which are DowJones30 companies and the other half non-Dow Jones 30 companies, the stock market noise for the firms with more liquid stock is estimated as having smaller volatility in market noise processes. In our second essay, the frailty idea described in Duffie, Eckner, Horel, and Saita (2009) is expanded to industry-specific terms. The MCEM algorithm is used to estimate parameters and random effect processes under the condition of unknown hidden paths and analytically-difficult likelihood functions. The estimate used in the study are based on U.S. public firms between 1990 and 2008. By introducing industry-specific hidden factors and assuming that they are random effects, a comparison is made of the relative scale of within- and between-industries correlations. A comparison study is also developed among a without-hidden-factor model, a common-hiddenfactor model, and our industry-specific common-factor model. The empirical results show that an industry-specific common factor is necessary for adjusting over- or under-estimation of default probabilities and over- or under-estimation of observed common factor effects. Our third essay combines and extends works of the first two essays by proposing a common model frame for both structural and intensity credit risk models. The common model frame combines the merits of several default correlation studies which are independently developed under each model setting. Following the work of Duffie, Eckner, Horel, and Saita (2009), we apply not only observed common factors, but also un-observed hidden factor to explain the correlated defaults. Bayesian techniques are used for estimation and generalized Gibbs sampling and Metropolis-Hasting (MH) algorithms are developed. More than a simple combination of two model approaches (structural and intensity models), we relax the assumptions of equal factor effect across entire firms in previous studies, instead adopting a random coefficients model. Also, a novelty of the approach lies in the fact that CDS and equity prices are used together for estimation. A simulation study shows that the posterior convergence is improved by adding CDS prices in estimation. Empirical results based on daily data of 125 companies comprising CDS.NA.IG13 in 2009 supports the necessity of such relaxations of assumption in previous studies. In order to demonstrate potential practical applications of the proposed framework, we derive the posterior distribution of CDX tranche prices. Our correlated structural model is successfully able to predict all the CDX tranche prices, but our correlated intensity model results suggests the need for further modification of the model. / Statistics

Recently, the financial world witnessed a series of major defaults by several institutions and investment banks. Therefore, it is not at all surprising that credit risk analysis have turned out to be one of the most important aspect among the finance community. As credit derivatives are long term instruments, it is affected by the changes in the market conditions. Thus, it is a appropriate to take into consideration the effects of the market economy. This thesis addresses some of the important issues in credit risk analysis in a regime switching market. The main contribution in this thesis are the followings: (1) We determine the price of default able bonds in a regime switching market for structural models with European type payoff. We use the method of quadratic hedging and minimal martingale measure to determine the defaultble bond prices. We also obtain hedging strategies and the corresponding residual risks in these models. The defaultable bond prices are obtained as solution to a system of PDEs (partial differential equations) with appropriate terminal and boundary conditions. We show the existence and uniqueness of the system of PDEs on an appropriate domain. (2) We carry out a similar analysis in a regime switching market for the reduced form models. We extend some of the existing models in the literature for correlated default timings. We price single-name and multi-name credit derivatives using our regime switching models. The prices are obtained as solution to a system of ODEs(ordinary differential equations) with appropriate terminal conditions. (3) The price of the credit derivatives in our regime switching models are obtained as solutions to a system of ODEs/PDEs subject to appropriate terminal and boundary conditions. We solve these ODEs/PDEs numerically and compare the relative behavior of the credit derivative prices with and without regime switching. We observe higher spread in our regime switching models. This resolves the low spread discrepancy that were prevalent in the classical structural models. We show further applications of our model by capturing important phenomena that arises frequently in the financial market. For instance, we model the business cycle, tight liquidity situations and the effects of firm restructuring. We indicate how our models may be extended to price various other credit derivatives.

L'elemento che distingue un'operazione di cartolarizzazione consiste, secondo la definizione espressa nell'accordo di Basilea2, nell'identificazione di almeno due differenti posizioni di rischio (tranche), stratificate e subordinate, emesse a fronte di uno specifico portafoglio di attività. Nonostante il ricorso al tranching sia ampiamente diffuso e standardizzato le determinanti che giustificano il ricorso all'emissione multi-tranche sono ad oggi poco approfondite. Inoltre, i titoli emessi a fronte di operazioni di cartolarizzazione (CDO) possiedono profili di esposizione al rischio di credito differenziati, in termini di incidenza delle perdite attese ed inattese, ed in termini di correlazione con altri fattori di rischio: la valutazione del profilo di rischio è condizione necessaria per l'attribuzione di un giudizio di rating e per la definizione di un appropriato premio al rischio (pricing). Si rivela necessaria tanto la stima della distribuzione delle perdite del portafoglio (credit risk modelling) quanto l'analisi strutturale dei flussi di cassa generati e l'allocazione degli stessi alle tranche (cash flow modelling). Sulla base della letteratura di security design la tesi intende valutare l'efficienza del processo di intermediazione basato sulla cartolarizzazione multi-tranche rispetto all'intermediazione bancaria tradizionale e a forme di asset-backed security con unica tranche e focalizza l'analisi attraverso una verifica empirica delle teorie economiche a supporto del tranching, con particolare riferimento alla cartolarizzazione dei prestiti concessi ad imprese di piccola e media dimensione, oggetto di analisi specifica condotta nell'ambito di un'esperienza di stage presso il Fondo europeo degli investimenti. Quindi, grazie alla realizzazione di un modello computazionale sviluppato con un software di pianificazione finanziaria multidimensionale (Quantrix), la tesi presenta un approfondimento delle technicalities, mediante una modellizzazione dei flussi e della loro allocazione (Waterfall Payment Order), allo scopo di apprezzare il valore aggiunto di queste strutture di intermediazione. Aspetto, questo, non sviluppato nella letteratura accademica. L'analisi si rivolge alle operazioni realizzate nell'ambito dei due principali programmi di cartolarizzazione dei prestiti alle PMI attuati in Europa (Ftpyme e Promise). / Securitisation is a structured finance instrument which involves pooling of financial assets (such as loans and bonds) and creating multiple tranched liabilities, collateralized debt obligation (CDO), of a single issuer with different risk-return characteristics, which are sold as separate securities. According to the New Basel Capital Accord, tranching is the key feature that distinguishes securitisation transactions; although commonly applied, the factors that determine the extent and the nature of tranching remain largely unknown. Moreover, because tranching allows the risk characteristics of the collateral pool to be transformed, it contributes to transaction complexity in assessing the risk properties of such structured instruments: the risk profile that can be generated through tranched exposure, in terms both of expected/unexpected incidence losses and correlated default of pool assets, can lead to substantial differences among tranches, depending on the level of subordination below a certain tranche. Key to the reliability of structured finance pricing and ratings is the accuracy in assessing the credit risk in the underlying portfolio (credit risk modelling), as well as the accurate modelling of the distribution of cash flows to different classes of CDO (cash flow modelling). By analyzing the finance literature relating to security design and securitization this thesis provides an analysis of the efficiency of financial intermediation model based on securitisation and an empirical test of theories supporting the economic added value of tranching, with regard to SMEs loan securitisation, which topic was specifically investigated during a stage at the European Investment Fund. By realization of a computational model, performed using a multidimensional modelling software (Quantrix), the thesis closely examines securitisation transaction's technicalities, by modelling both portfolio cash flows and funds allocation (Waterfall Payment Order), in order to asses the ability of the structure to withstand various stressed scenarios. This analysis offers an analytical and micro-approach to securitisation transactions, which has not deeply investigated in academic literature yet. The model applies to SMEs loan securitisation transactions, concluded within specific securitisation European Programme (Ftpyme in Spain and Promise in Germany).