Hi all,I want to run an analysis to compare VaR computed using zero coupon rates vs using spot rates on a swap portfolio. Because the analysis results will greatly depend on the portfolio content and also because i'm fairly new to the concept of VaR, i wanted to get your expert opinions on how you think VaR will behave on these two scenarios.Here's my first interpretation: as long as the spot curve is flat, there wont be much difference between spot rates and ZC rates movements, hence little difference between the two VaR. However, on a steep curve, high yield environment, the variation of the ZC rates vs spot rates will be exponential. leading to a greater VaR number using ZC rates vs spot rates. This of course if the portfolio contains long dated positions on high yielding currencies. Any qualitative opinions, personal experience, on using ZC rates to calculate VaR are also very much welcome. thanks

Your estimate of VaR (or, for that matter, any other quantity) does not depend on how you choose to express your term structure of interest rates, but rather on how you assume that the term structure moves. And, at least in my neck of the woods, "spot rate" and "zero coupon rate" actually mean the same thing. Maybe you mean "par rate" vs "zero coupon rate"? Any given scenario expressed in terms of one set of rates will map into a scenario in terms of the other.

yes i mean par rates by spot rates. so you are saying the shape of the curve has no influence?

bearish is not suggesting that the shape of the curve has no influence on VaR. Par and zero curves contain the same information about the shape of the curve, one is merely a transform of the other.