Enhancement #1
Separated Italy CMT into only two curves: Low (coupon <4%) & High (coupon >=4%) for more accurate trade discovery and rolldown analysis when analyzing BTPS bonds against the CMT curve.
Previously, we separated the low coupon curve into two curves. However, we found having three separate curves didn’t guarantee enough liquid points.
As of this week’s release, the low coupon curve fits coupons of less than 4%. This curve will naturally converge as an “ultra low” curve over time as the ultra low coupons will continue getting supplied, and will have an increasingly larger influence based on our default weight (Issue Size * DV01).
IT CMT Low fits all bonds with a coupon less than 4% (“CMTITLO#” in Sprd/Bfly & Forward Swap Matrix)
IT CMT High fits all bonds with a coupon of 4% or greater (“CMTITHI#” in Sprd/Bfly & Forward Swap Matrix)
Enhancement #2
Refined the USD CMT curve to fit the original issue within each sector to create a smoother curve for more powerful relative value analysis
For the USD CMT curve, we found that using all the notes and bonds in that sector to build the curve resulted in some kinks and dislocations largely because of the coupon effect and liquidity differences between the original issues. We instead choose to build the USD CMT curve using only the original issues for each sector. Specifically:
For 0-2y of the CMT curve, we only use original issue 2 year notes
For 2-3y of the CMT curve, we only use original issue 3 year notes
For 3-5y of the CMT curve, we only use original issue 5 year notes
For 5-7y of the CMT curve, we only use original issue 7 year notes
For 7-10y of the CMT curve, we only use original issue 10 year notes
For 10-30y of the CMT curve, we only use original issue 30 year bonds
Enhancement #3
Plot the Par Rate curve, instead of the Spot Rate
Par Rate is defined as the bond’s par coupon, which is the coupon such that the curve discounted clean PV of the bond cashflow equals par.