- Loan Portfolio
- 2nd Edition
- Credit Portfolio Modeling Handbook
- Solutions | Expersoft
- Credit Portfolio Modeling Handbook Credit Suisse.pdf -...
In my opinion, the efforts of risk measurement are vanity without promoting the measurement into management. In summary, sound credit risk management must be transparent and well perceived. After all, banking business is a cyclical business with frequent expected loss and threatening unexpected losses. I could point out that the most opaque black box in many financial institutions is the quality of credit asset. The figures that bank provided in the annual report only depicts the past and have no value regarding the future asset quality. The disclosure of economic capital will give outsiders a clue.
Also is an action of risk governance that will distinguish bank from her peers. Any error and unintentional deviation from the best practices remain my own responsibility. Eric Kuo, Sep, 7. Table of content Section 1: Foreword and Introduction Performance metrics The Vasicek formula Correlation estimation: R The expected loss Maturity adjustment The last decade has seen the development of models to compute portfolio credit losses for bonds and loan portfolios.
The important output from the credit portfolio model is so called economic capital which is used to gauge how many amount of potential unexpected loss a bank is exposed to given the current credit portfolio constitution. Although, the Basle 2 has taken the first steps to amend the capital requirement and to promote banks to implement the internal credit risk models for better estimating the unexpected loss regulatory capital. In the BIS regulatory model, the potential exposures are given by an add-on factor multiplying the notional of each transaction.
It is simple to implement, but the model has been widely criticized because it does not accurately capture the diversification effect and concentration risk of portfolio. By contrast, credit portfolio models measure credit economic capital and are specifically designed to capture the portfolio effects, specifically obligor correlations. Although superficially they appear quite different—the models differ in their distributional assumptions, restrictions, calibration and solution1. The major limitations, in my point of view, are: the vendor models are expensive and complexity.
Expensive means the subscriber needs to pay for the software expense each year, unless bank has a big commitment on the use of economic capital. Most of the model comprises sophisticated mathematic modeling and are difficult to explain in a simple spreadsheet. Both of the above are the motivation of this document. In addition, it is vital for banks to estimate the unexpected loss of their credit portfolio to better understand the uncertainty. In the following, this document will explain the method that grounded on this simple credit portfolio model.
The limitation of this model and directions of future improvement are also discussed.
I also briefly introduce the management applications and the concept of active credit portfolio management in this document as well. To prevent from the insolvency and result in financial crisis, banks need to reserve a certain amount of capital to protect from unexpected loss except for the provision reserve. If the regulator takes a conservative stance in the regulatory capital policy, then bank needs to hold billion for this billion of loan portfolio.
However, the occurrence of this maximum loss is close to zero. The event implies that all obligors are going to be insolvent in the same time. The essence of credit portfolio management is to establish portfolio balance with adequate diversification2. This mitigates the consequences of the portfolio's volatility of value sometimes termed unexpected losses to a level where an institution can survive such losses given its reserves and capital. The Basel committee, therefore, investigated the existed credit portfolio models and assistance from the best practices3.
Page 5, page 6. Restricted Bank capital is reserved as a cushion to absorb unexpected loss. Target 1. It might over or EL under estimated the Time Probability risk. The capital requirement is reserved for the losses that exceed the expected loss.
It might over or under estimated the risk. The exposure may be different: for example the average credit loss is out of 1, of exposure, while as the exposure may be expand to 2, To compare the absolute amount of expected loss may not be appropriate. This implies that there is only 0. UL more important than EL small number 1.
- IFRS 9 and CECL Credit Risk Modelling and Validation - 1st Edition?
- Reconfigurable Computing: Architectures, Tools and Applications: 8th International Symposium, ARC 2012, Hong Kong, China, March 19-23, 2012. Proceedings.
- Risk Management Toolbox.
- Destroying America - The Real Issues Politicians Avoid;
- Advanced Techniques in Computing Sciences and Software Engineering.
EL more important than UL large number of of relatively good quality loans loans minimises Impact of fluctuations 2. High correlation 2. Low correlation 3. Significant capital requirements 3.
Credit Portfolio Modeling Handbook
Based on the equation; the asset correlation for the AA grade8 is around Correlations between these boundaries are modeled by an exponential weighting function that displays the dependency on PD. The upper and lower bounds for the correlations and the functions are based on the empirical studies. The reason that the retail products have lower 8 PD of AA grade is 0.
PD of B- grade is Several studies also confirmed the same result9. On the other hand, the mortgage has higher dependency with the real estate industry and is deeply influenced by the economy; therefore, the mortgage has higher asset correlation compare with other retail products Therefore, they turn to implement the correlation into the probability of default.
The rational is that better rating obligor usually has larger asset size; larger asset usually has a higher dependency with the state of economy Therefore, if the economic declines, the Honhai company will be easier influenced by the economic downturn than SMEs or retail products may have. The empirical relationship between average asset correlation,firm probability of default and asset size.
Restricted Same lending amount, different capital charge are result from correlation. As a result, the EL is the same for both exposures.
Solutions | Expersoft
Given the correlation formula, the corporate client has an Even though, the correlation difference between this corporate client and mortgage is merely 3. Probability of loss The maximum amount of loss is the total principle, in our case it is million. If the loan is still performing, the bank needs to reserve 0. In oppose to the corporate loan, the mortgage also needs to reserve the same provision, but the capital is far lower. The amount that is not covered by the EL and UL is so called tail risk.
- How We Lived Then: History of Everyday Life During the Second World War, A.
- Credit portfolio management;
- About AIWMI.
- tomiddperspafi.gq The web's biggest credit risk modeling resource.;
The tail risk is a risk that rarely happens but once it does, it will cost you an arm and a leg. The recent sub-prime credit lesson is a perfect example to demonstrate the importance of tail risk management. Differences between AIRB capital and Economic Capital There are five major differences between economic capital and regulatory capital, in my point of views: 1. The X axis represents for the amount of credit losses.
Credit Portfolio Modeling Handbook Credit Suisse.pdf -...
On the other hand, the Y axis stands for the occurrences of the corresponding credit losses. Better rating grade means bank need to hold more capital for protecting unexpected loss. The economic capital represents for risk governance of a bank. As shown in the chart below: the Winterthur illustrates that their risk exposure in line with risk taking capacity to a confidence of The The 0.
The possibility of loss amount exceeds the current available capital is 0. The consideration of diversification effect, which usually refers to the estimation of customized asset correlation, instead of using the constant correlation suggested by the Basel.
Download a free trial. Download now. Stress Testing Perform stress testing and sensitivity analysis on financial portfolios. Lifetime probability of default for a stress test. Regulatory capital by asset class. Credit Risk Modeling Model and analyze the risk exposure of credit portfolios. Credit Scorecards Modeling Use the Binning Explorer app to develop credit scorecards by applying auto-binning algorithms or interactively adjusting edges, merging bins, and splitting bins.
Binning Explorer app for credit scorecard modeling. Credit Risk Simulation Perform copula simulations based on probability of default or credit rating migration to analyze the risk of credit portfolios. Modeling Correlated Defaults with Copulas. Portfolio losses based on copula simulations. Risk Parameters Estimation Estimate probability of default PD using various methods, including structural models, reduced-from models, historical credit rating migration, and other statistical approaches. Lorenz curve for representing the distribution of risk exposure. Results from multiple VaR backtesting models.
Expected Shortfall Backtesting Backtesting models for expected shortfall ES include conditional test, unconditional test, and quantile test. Improve your ability to identify which customers are most profitable with an all-in-one assessment of their creditworthiness. Decisioning As A Service is a streamlined delivery method engineered to help Experian clients gain greater value from data and decisioning products.
It does so by providing flexible, real-time access to more data sources, attributes, scores and analytics. Differentiate the highest spenders to deliver better service, provide superior rewards, and even offer significantly higher credit lines. A new generation of category-specific campaign targeting and tailoring tools for financial services marketers seeking to improve the performance of their campaigns.
Financial Personalities is the only campaign targeting and tailoring system of its kind in the marketplace today.