They cannot be coputed directly. It is a static model. You can estimate the Greeks by shifting the market-data inputs, but there are no analytic formulas. Since most firms calculate Greeks by shifting data, regardless of the availability of analytic formulas, this is not a bad possibility. The difficult part is determining which sensitivities you may want to hedge against.
Chat with our AI personalities
While compound correlation is the correlation found by calibrating the Gaussian copula model to the price of a CDO tranche (for example 3-6%), base correlation is found by calibrating to the price of a first loss tranche, i.e. to the sum of all tranches up to an attachment point (for example 0-6%, the sum of 0-3% and 3-6%). The curve of correlations obtained by calibrating to first loss tranches turns out to be much smoother and more stable than that obtained by calibrating to plain tranches.
I think you mean cupola which is latin for 'small cup' and is a cup-like structure that crests belltowers, dome, etc. Usually found on churches because of a law that should make the church a highest building in a town. In order to get the extra height, a cupola is usually built.