The key idea in valuing a CDS is a fair deal: the (probability-adjusted) expected PAYMENTS (i.e., made by protection buyer) should equal the expected PAYOFF (contingent, made by seller)
ABX introduced a means for the transparent pricing of subprime risk (where previously there was none). In the second part of this briefcast, I show how the authors instructively calculate the implied spread given the index price.
In a securitization, we can take a balance-sheet perspective: on the left-hand side, credit-sensitive assets (collateral) have value, create cash flow, and contain risk; on the right, liabilities (tranches issued to investors) IN TOTAL must preserve value, cash flow and risk.
The next building block is mapping transitional probabilities to standard normal variables; then using a bivariate normal to capture joint probabilities of default
How the RSS is calculated (test of FLV format)
This is a review which follows Jorion's (Chapter 7) calculation of marginal value at risk (marginal VaR). Marginal VaR requires that we calculate the beta of a position with respect to the portfolio.
08:29
Loss distribution for credit enhancement in securitization
This continues to follow the subprime securitization case study by Aschraft. There are two steps: 1. Specify the loss distribution; and 2. Map the target credit rating to the implied credit enhancement
A review of the method used in the first building block of CreditMetrics, a ratings-based credit risk portfolio model
07:28
Beta distribution for loss given default (LGD)
The beta distribution is typically used for modeling loss given default (1 - recovery rate).
A brief review of Crouhy’s approach to setting the internal capital charge for market risk. Internal means to distinguish from regulatory (external) capital requirements such as Basel II.
A simple (two-variable) regression has three standard errors: one for each coefficient (slope, intercept) and one for the predicted Y (standard error of regression). While the population regression function (PRF) is singular, sample regression functions (SRF) are plural. Each sample produces a (slightly?) different SRF. So, the coefficients exhibit dispersion (sampling distribution). The standard error is the measure of this dispersion: it is the standard deviation of the coefficient.
The best hedge is based on portfolio volatility in the mean-variance framework. Specifically, 1. Given a current portfolio with value (W), and 2. Given an asset (A) with correlation (rho) to the portfolio, 3. What is the trade that produces the minimum volatility for the new portfolio (W+a)?
In MG, the underlyings were short positions in long-term forward contracts to deliver oil. The hedge was a stack-and-roll hedge: long positions in short-term futures contracts that were rolled over consecutively. The strategy depended on the continuation of (i) stable or gently increasing spot oil prices and (ii) backwardation
Review of key players including special originator, purpose entity, custodian, underwriter, investors, legal, and credit rating agencies
Liquidity adjusted value at risk (LVaR)adjusts (increases) VaR as a function of the bid-ask spread.
RAROC is a risk-adjusted performance measure (RAPM): risk-adjusted return divided by economic capital (i.e., the capital reserved to cover unexpected losses).
08:20
Risk contribution of credit to portfolio unexpected loss
Risk contribution is analogous to systematic risk in single-factor (capital asset pricing model): as Ong says, it is a measure of the âundiversified risk of an asset in the portfolio. It is the amount of credit risk which cannot be diversified away by placing the asset in the portfolio.â
A securitization is a structured finance with three ingredients: 1. Pooled credit-sensitive assets; 2. Transfer of credit risk; 3. Tranched liabilities
The capital market line is determined by a mix of: the riskfree asset and the market portfolio. The market portfolio, in turn, consists of all risky assets (this example has only two assets).
07:58
Expected loss (EL) on credit asset if PD, LGD are correlated
Expected loss (EL) calculations typically assume no correlation (i.e., they assume independence) between probability of default (PD) and loss given default (LGD). Basel II internal ratings-based (IRB) approach to a capital charge assumes independence between PD & LGD. How can we compute expected loss (EL) if there is correlation between EDF/PD and LGD/recovery?
Not finding what you want? View results from YouTube.
Comments