
Radon–Nikodym Derivative see Equivalence of Probability Measures
Random Factor Loading Model (for Portfolio Credit)
Consider a portfolio of N risky assets, all assumed (for simplicity) to generate a $1 loss at the time of default. Let τi denote the random default time for asset i, such that the total portfolio loss L(T) on the horizon [0, T] is
![]()
From credit default swap (CDS) or bond markets, we can normally extract risk-neutral survival probabilities
![]()
for all T; this information locks risk-neutral expected portfolio losses at
![]()
To be able to construct the entire distribution of L(T)—and not just its first moment—we need additional information about the default codependencies among the N assets.
The default codependence model that we consider here is in the class of factor models, in the sense that codependence is induced solely by a scalara random variable Z—the so-called systematic factor—that affects all assets through a factor “loading” function. Conditional on Z, all N default times τi are assumed to be independent of each other.
In practice, specification of factor loading is done ...
Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.
Read now
Unlock full access