Curious if someone could help me out with pricing this trade idea, or just give me some general tips on a direction I need to head to go about this. I attached a photo if to see how I set up the idea so far in excel, the specs of it. Thanksenter image description here

  • $\begingroup$ What's the question? 5s30s is directly observable e.g. USYC5Y30 Index on BBG. $\endgroup$
    – user42108
    Oct 12, 2021 at 21:14

1 Answer 1


Not sure what you mean by pricing this trade since the price of a future is given by the exchange. You can get bond futures data for free from CME (TU is the symbol for the 2y and WN for the 30y). I’ll assume you’re asking about weighting the legs. There are many ways to do it but here are some common ones:

  1. Equal weights: +1 on TU and -1 on WN. This is the most trivial case. You’ll buy/sell an equal number of contracts but risk overexposing to the longer dated future with higher duration.
  2. DV01 weighted: to neutralize the delta or duration you can calculate how many futures are needed such that the net risk is zero. For example, buy 10 contracts of WN, how many TU are needed (around 85)? The risk of a future is the DV01 of the underlying cheapest-to-deliver divided by the conversion factor. You will most likely need a pricer for that such as Bloomberg or Eikon. I would say this is the most common approach.
  3. PCA weighted: you can perform principal components analysis on the yield curve and use the loadings as weights for your trade. This is useful because you isolate the main risk of this trade: steepening/flattening of the curve which is reflected by the second principal component (or eigen value). You could weight the legs such that the first PC is minimized (i.e. changes in levels not steepness). This paper gives a great introduction.

There are of course other methods as well, e.g. minimizing VaR or neutralizing beta. Ultimately it depends on what your risk goals are. Great answers on this topic can also be found here and here.

  • $\begingroup$ Great, that helps! Really appreciate the response. $\endgroup$ Oct 12, 2021 at 21:26

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