Variable Maturity Giant First Loss

A variable maturity GiantFirstLoss trade has a non-vanilla collateral debt obligation (CDO) structure, in which the maturities of the obligors in the underlying collateral pool can be different from the trade maturity.

The valuation model serves the purpose of pricing a variable maturity GiantFirstLoss trade. The trade has a non-vanilla collateral debt obligation (CDO) structure, in which the maturities of the underlying obligors could be different from that of the CDO trade

The correlated defaulted events are generated by the Monte Carlo (MC) simulation via either normal copula model or Poisson model. At the present time, the approved default correlation model is the normal copula model.

The losses associated with the defaulted obligors are allocated to each tranche according to the subordination of the tranche. In the variable maturity GiantFirstLoss trade as well as this loss claiming scheme, a pay down scheme of the tranches, associated with the reduction of the collateral pool due to the default events and the early retirement events of the obligors, is implemented.

The sensitivities are calculated using weighted MC method. Currently SH3 supports the computation of WMC credit spread sensitivity, WMC default sensitivity, WMC interest rate sensitivity, recovery rate sensitivity, and carry.

The model has certain changes made to the payoff functions of the approved GiantFirstLoss pricing template. Both the default correlation model and sensitivity computation methods remain unchanged.

The vanilla CDO model, named GiantFirstLoss trade, is employed as the benchmark. If we switch off the two non-vanilla features, the variable maturity GiantFirstLoss trade will become a vanilla CDO. The default pay down feature could be switched off by setting the recovery rate to zero and the maturity pay down feature could be turned off by forcing the maturities of the obligors in the collateral pool to be equal to or larger than the trade maturity.

• valuation models
• risk models