Many dealers are moving into trading of Constant Maturity Default SwapsDoes anyone have some insights about which approach they use to price such deals?Even if they use some stochastic intensity based approach, calibration & hedging will become a big issue, due to lack of liquidity in Default Swaptions market.Any ideas will be greatly appreciated....

The pricing difficulty is with jumps to default, especially if they occur relatively early in the life of the deal (risky cashflows are typically discounted fast enough for these things that jumps occurring more than a couple years out have minimal impact on day 1 valuation.)Now if you have option (or equiv portfolio of options to CMDS) to sell protection, you can be pretty sure it will be out of the money given spread jump, therefore you can treat spreads (or hazards) as diffusion process and get a decent approximation to price. For option to buy protection, this is not true, and while I have seen some proposed pricing methodologies, I have yet to see anything that is both accurate and tractable for maturities longer than around 6 months. (Short maturities can be approximated by lognormal diffusion with a convexity adjustment to account for non-diffusion properties of spread evolution.)Don't know if that helps.

Hi Jabairu,We might expect that dealers will promote both (CDS options & CMDS) in order to hedge each other in a some way....??????It seems to me strange making convexity adjustment for short maturities. Actually convexity adjustement becomes an issue for longer maturities (the longer is the underlying maturity the higher is the adjustement....)May be an affine-jump diffusion approach of intensity rate will sort out the problem?! It would be possible to calibrate it to the volatility skew...

that's the whole point - the convexity adjustment is small for short maturities, therefore we can justify our assumption that it is reasonably close to the true error caused by non-diffusion terms in the underlying process.jump diffusion intensity processes are great, in theory, but there is no volatility surface to use for calibration. trading is very thing, and the only vols that actually might actually get traded in the near future will be ATM options. If I am a dealer, I am going to be scared to trade OTM options because if the counterparty does a trade with me, I can be pretty sure that I under-estimated vol. As you can see, this becomes more a concern as maturity extends.

Could you please tell me how the deal get structured for CM DFS? The protection seller still pays 1-R if R is the recovery rate upon default. How about the buyer? I have seen many DFS deal that we called floaters where the premium pay 3M Libor + spread. Is this related to the this kind of CM DFS trades?Appreciate your response.Yeren

what is CMDS ? Actually, i also don't 100% understand a CM Swap. can anybody explain rougly ?

point of cmds is to trade the spread without the credit carry... i.e. I can quote 5y cds all day long, but after I do a trade, I'll have to quote 4.99y cds as well... seems a bit inconvenient from a mathematical pov, right? probably people push this as a way to get set up for spread options, for the same reason...and for banks who made loans to companies, there isn't any reason for their hedge to expire - the only reason would be for 'real leverage dynamics', which aren't that well understood. What I mean is that maybe the company pays down their loan, but they do that by issuing bonds most of the time (i.e. the bank has the opportunity to 'reinvest' in their spread exposure). But you see my point - the real CMDS spread driver will be things like going IPO, company takes a big hit to equity, new leveraging with CP or bond issuance, and not market technicals. but still I don't see a big market. What you read when you first studied CDS was that it was a silver bullet. It wasn't (but that's not advertised!) People are still trying to figure out what the right thing to trade really is. Still struggling away, yes..