Formula for extending LIBOR zero curve

[hsuwang]

Hello David,
for the formula on page 46 of your screencast: Ri+1 = (Fi(Ti+1 - Ti) + RiTi)/Ti+1
Is there a more intuitive way to better understand this formula or is memorization the more preferable way?

I was trying to see if there's a relationship between this formula (where we extract zero rates from forward rates) and the formula that we used in chapter 4: (R2T2 - R1T1) / T2-T1 , where we were trying to extract forward rates from spot rates, I was trying to see if I can get one formula from the other, but wasn't really able to do so.

- another question is that when I use the (R2T2 - R1T1) / T2-T1 formula to try to figure out the forward rate by the spot rates provided on page 46 (just for testing), I got a forward rate of 5.3%: Rf= 4.893%+(4.893%-4.8%)*(400/91), but the table shows that the forward rate is instead 5.5%. Can you please tell me what I'm doing wrong here? Thanks!

 

[SURAJM]

Jack,

you got the correct answer .... just that you are looking at the wrong period .... the forward rate that you have calculated is for the period of 400 days to 491 days which is indeed 5.3% as per the table on pg 46

 

[David Harper CFA FRM]

Surajm: thank you

Jack: I don't think my table is especially clear (frankly, I regret the schedule is so aggressive because normally I would/should do a better service of sharing this key building block). This is such a key building block that, in my point, memorization is *not* useful (i'd go further, i think trying for memorization here can actually contribute to error).

I just started a sticky thread, my thought was to keep a living thread of the key building blocks that will inform the reviews, etc.

So, in regard to "Markets & Products," extracting implied forwards (or vice-versa - extending the spot from forwards) is highly thematic.

So please see this thread that I just started (and XLS): key building blocks; entry here means it will re-appear in the review.

Thanks, David