| Investment Knowledge |
 |
 |
 |
 |
|
Basket Default Swap A pricing model is presented to calculate Mark-to-Market (MTM) and all sensitivities for basket default swaps and Collateral Debt Obligations (CDOs) (FirstNofM, GiantFirstLoss, GiantFirstLossPayEnd, Caribou, and Reindeer). It is composed of the credit library, BulkCurveGenerator, five outstanding pricing templates, and Scenario Manager. Within the modeling framework, the sensitivities measure the change of present value (PV) when certain risk factor is perturbed. For example, recovery rate sensitivity is calculated by perturbing the recovery rate for each obligor in the reference collateral pool of the trade, which can be defined as follow. In spreadsheet model the credit spread sensitivity is defined as the change of PV when the credit spread curve is parallel shocked several basis points for each obligor in the collateral pool. The default sensitivity is calculated by moving the default time of perturbed obligor to the valuation date to reflect the default risk of the obligors in the collateral pool. The correlation sensitivity is usually measured by changing the flat correlation of the collateral pool ten percent. 1-day theta and 6-month theta is defined in model as moving valuation date one day and 6 months forward, respectively, with all other parameters unchanged. The default sensitivity and the credit spread sensitivity have been vetted before. Except for correlation sensitivity, in SH3 there exist two versions of risk measures for all sensitivity tests. One version is the regular way by perturbing the risk factor and recalculating PV. The sensitivity is then calculated by Eq.(1) directly. The other way, which is more efficient, employs a re-weight approximation. It is denoted as weighted Monte Carlo (WMC) sensitivities. There is one major change to the credit library. An entropy threshold, which equals 10, is set in the library. It has been known that the WMC simulation sometimes would converge to an apparently “wrong” value. An entropy threshold is thus set to identify a possible “wrong” solution. Apart from managing the computation of MTM and sensitivities using those pricing templates, Scenario Manager (SM) calculates the sensitivities through regular method. That is, by loading the pricing templates, the unperturbed and perturbed PV for a trade are calculated and then the sensitivity is defined as the difference of the two values. The risk measures built in SM are recovery rate sensitivity, interest rate sensitivity, correlation sensitivity, and 1-day and 6-month theta without re-weigh approximation.
|