Serving the Quantitative Finance Community

 
hiroshima
Topic Author
Posts: 3
Joined: August 24th, 2018, 12:42 pm

CVA using Hull&White single factor model

August 24th, 2018, 1:52 pm

Hello !

I am now working on the calculation of the CVA. To do this I need to make a Hull&White single factor model on Matlab that will enable me to compute the exposures. In HW model I have 10 dates and I simulate 100 trajectories and I had some questions:
1)is it normal if each trajectory is totally random and does not look like an interest rate curve ? 
2)from my HW model, how do you suggest me to compute the exposures ? 

Thank you very much !
 
User avatar
bearish
Posts: 5186
Joined: February 3rd, 2011, 2:19 pm

Re: CVA using Hull&White single factor model

August 25th, 2018, 3:48 am

I haven’t the faintest idea, but have a couple of observations, questions and recommendations. First, this question is likely to get a better reception in the Student forum. Second, the one factor HW model is a terrible choice from an empirical point of view. Third, a hundred paths is a shockingly low number. What instruments/positions are you trying to calc CVA for? How are you accounting for the default risk? What do you actually mean when you say that your trajectory doesn’t look like an interest rate curve? Should it? A much more detailed post might result in more helpful comments (or ridicule, but that’s the usual risk inherent in asking questions).

Edit: never mind the first suggestion. Change that to read: never post the same question in multiple forums.