2
$\begingroup$

I was wondering how to calculate the forward rates based on OIS discounting for the half year terms. I know how to do this for the full year terms -> just making sure that the two legs are equal to each other. The problem is that I don't have fixed payments on the half year terms. For example when one wants to calculate the 1,5 OIS discounted forward rate you only have 1 fixed payment versus 3 floating payments. How to deal with this? One method which can be used is just calculating the full year rates and then interpolate between these rates, but I wonder if there is a method which can calculate these rates without using interpolation?

Next to that what does a 1,5 swap rate exactly mean, is this just an interpolated rate because you will have the same problem here when there isn't a payment on the half years.

$\endgroup$
2
  • $\begingroup$ What do you mean "you only have 1 fixed payment versus 3 floating payments". I have never heard about this new kind of swap. $\endgroup$
    – nbbo2
    Jan 14, 2016 at 11:32
  • 1
    $\begingroup$ I don't think this kind of swap is really common but it's more a fake instrument to calculate the forward rates based on OIS discounting. $\endgroup$
    – Oamriotn
    Jan 14, 2016 at 13:13

1 Answer 1

3
$\begingroup$

Which currency are you looking at ? Say that your 1y swap would have yearly fixed payments vs 3M floating payments. Your 1.5y swap would probably have:

  • a fixed payment 6m after effective date and another fixed payment 18m after effective date
  • regular quarterly floating payments

Your curve was built with 1y and 2y swaps, nothing in the middle ? Then yes, your interpolation choice would matter a lot. I would look at 3m forward rates and make sure that they optically make sense.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.