DiscountCurve不知道QuantLib python中的评估日期

Question

I am using quantlib in python. In order to construct a DiscountCurve object, I need to pass a vector of Dates and corresponding discount factors. The problem is that, when I change the evaluation date to account for settlement days, the curve object is not shifted/adjusted properly and the NPV of the bond does not change as a function of evaluation date.

Is there any way around this? Do I have to construct a different DiscountCurve by shifting the dates whenever I change the number of settlement days?

Ideally, instead of passing a vector of dates, I should be able to pass a vector of distances between consecutive dates but the very first date should be allowed to be the evaluation date.

Answer 1

No, unfortunately there's no way around this. For this particular class, you'll have to recreate an instance when your settlement date changes.

Writing a version of the class that takes distances between dates can be done, but it's not currently available. If you write it, please consider creating a pull request for inclusion in the library.