Biases can be introduced into the computation of duration of nondefault-free securities when the time pattern of losses from default is not explicitly taken into account.
Biases can be introduced into the computation of duration of nondefault-free securities when the time pattern of losses from default is not explicitly taken into account. This monograph shows that measures of duration for fixed-income securities should account for the possibility that the promised cash flows will change in the event of an unexpected default by the issuer.