CDO pricing using single factor \(\mathcal M_{G-\mathcal{NI}G}\) copula model with stochastic correlation and random factor loading (Q2517674)

To observe the repackaged portfolios of the tranches of the credit risks of default losses of the collateralized debt obligation derivatives, a copula model \(\mathcal{M_{G-NIG}}\) of the mixture of the Gaussian distribution and the normal inverse Gaussian distribution is suggested in this paper. In two cases of stochastic correlation and random factor loading, the characteristic function of the default probability of the \textit{i}th entity is reduced in a way of somehow easier to do simulating. However, the data fitting verification is not provided. Reviewer's remark: In reading this paper, we would like to add two points: 1. Some misprinting appears in the line 13 of page 76, where all the tildes in the distribution of \(\epsilon_i\) should be deleted. 2. In the line 14 from the bottom of page 74, \(A_i\) actually depend on \textit{t} [cf. \textit{J. Hull} and \textit{A. White}, Valuation of a CDO and an \(n\)-th to default CDS without Monte Carlo simulation, J. Derivatives 2, 8--23 (2004)].