CBOT 2s/5s/10s fly

Y

Ye Olde Refco

Anybody know the 2s/5s/10s fly price/leg ratios?

I've seen the one that NAVEEVIa kindly put up, but I was wondering if any one had any other suggestions.

I've tried adding the 2's/5s with the 5's/10's but feel as though this is wrong.

All suggestions welcome!!!
 
Don't trade the yield curve w/o doing your own calculation of PV01 equality on the fly legs. IOW, don't use an arbitrary ratio.
 
Cheers Atticus, this is exactly what i'm looking for. But I have no idea as to how to calculate the PV01 equality on the fly legs. Could you help shed some light on this for me.

Cheers
 
Using Z9 contracts at curr mkt prices:
1000 2y contracts = 861 5y contracts = 598 10y contracts.

You know what to do with these to do a fly.

The calculation of instantaneous DV01 (assuming a constant CTD) is quite simple. I think there was a thread somewhere here where I attached some slides from Eurex about the calculation of contract DV01s.
 
I have also seen this done w/ linear regression weights instead of theoretical DV01s.

I haven't traded either so I can't say which is better for sure but regression weighing seems more logical to me. DV01 weighting assumes a parallel shift in the curve. Regression weights take into consideration the different volatilies along the curve and recent correlations.

But if you are trading a small number of contracts it might be the same answer anyway.
 
Thanks for all of the suggestions so far. I have been quickly enlightened.

Still, I have the problem of charting the fly on CQG. Does anyone have any good suggestions on how to translate the fly on to the chart?
 
Quote from black diamond:

I have also seen this done w/ linear regression weights instead of theoretical DV01s.

I haven't traded either so I can't say which is better for sure but regression weighing seems more logical to me. DV01 weighting assumes a parallel shift in the curve. Regression weights take into consideration the different volatilies along the curve and recent correlations.

But if you are trading a small number of contracts it might be the same answer anyway.
More...
There's a whole variety of ways to approach this. Tuckman suggests a few different methodologies, but it's by far not an exhaustive list.

Problem with regression-weighted flies is, as usual, the choice of the sampling period and the stability of your coefficients.
 
You must log in or register to reply here.
Back
Top