Can futures rolling affect curve constructing? Let's say that i'm using future 1 and 2 to construct the short end of my curve. As i understand it, rolling will create volatility (as volumes spike), which can affect the futures prices and thus the rates deduced from those futures. So is there a risk that the part of the curve implied from these futures presents some kind of jumps in this rolling period? Thank in advance.

| improve this question | | | | |

Futures roll can potentially create some short-term imbalances that distort market pricing. Whether or not that should be accounted for in curve construction is philosophical. In my mind, there's no reason to – you're building a market-based curve, and you shouldn't override the market. You can, however, build a separate "fair value" curve which incorporates some of your personal views on where things should be trading at. Comparing this fair value curve with the market-based curve is how you discover relative value trading opportunities.

Also, I wouldn't worry about "jumps" in the curve in general. There's no reason why the yield curve has to be perfectly smooth. You may have read about the humps in the T-bill curve around October maturity. The hump simply reflects market concerns about whether or not the debt ceiling will be lifted in time. There are always reasons for why these abnormalities exist; it's best to show them so as to reveal potential trading opportunities, rather than smoothing them out just to make the curve look nicer...

| improve this answer | | | | |
  • $\begingroup$ I understand your point. However, when pricing some "more complex" derivatives, smoothness is soemthing we're looking for. Don't you agree? $\endgroup$ – ababoua Aug 22 '17 at 8:16

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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