SERVING THE QUANTITATIVE FINANCE COMMUNITY

 
User avatar
Herd
Topic Author
Posts: 562
Joined: October 2nd, 2003, 12:48 pm

Is counterparty risk the new smile?

January 23rd, 2013, 9:14 am

Before everybody was obsessed with fitting the smile, Heston calibration and so on....And now, all of a sudden, this obsession has disappeared.It has been replaced by a new one, maybe even bigger: counterparty risk.How long is it gonna last for? What will be the next one? And what happened to the smile obsession?
 
User avatar
quartz
Posts: 427
Joined: June 28th, 2005, 12:33 pm

Is counterparty risk the new smile?

January 23rd, 2013, 11:50 am

Once it became clear that everyone was fitting his own personal parameter to the smile it was also natural to infer the value of such "advances"... (ok ok...)On the other hand CCR was always there even if under some radars, and is linked to both core banking and wars, both of which will never go away (but they'll change shape). Btw dont forget the copula frenzy.
 
User avatar
Herd
Topic Author
Posts: 562
Joined: October 2nd, 2003, 12:48 pm

Is counterparty risk the new smile?

January 23rd, 2013, 3:13 pm

yes, maybe the smile frenzy is closer to the copula one actually in the sense that counterparty risk exists outside models.copula is having fun with a simple model, whereas smile modeling is having fun with complicated models.
 
User avatar
Gamal
Posts: 2362
Joined: February 26th, 2004, 8:41 am

Is counterparty risk the new smile?

January 23rd, 2013, 3:31 pm

QuoteOriginally posted by: HerdBefore everybody was obsessed with fitting the smile, Heston calibration and so on....And now, all of a sudden, this obsession has disappeared.It has been replaced by a new one, maybe even bigger: counterparty risk.How long is it gonna last for?Not before I retire.
 
User avatar
PvalAnal85
Posts: 37
Joined: May 26th, 2009, 6:23 pm

Is counterparty risk the new smile?

January 23rd, 2013, 4:17 pm

Hi,I'm not so sure that Heston-parameter smile-fit modelling ever really went away. What we've seen is perhaps a vast unwind in the kind of complex structures (due to financial crisis) that made Heston calibration necessary in the first place. In addition, as models become "canonical", they are implemented at banks and hedge funds, and there is less of a need to discuss theoretical foundations of the model (as they're established in practice).Counterparty risk is the huge thing now because of Dodd Frank and Basel III. The breakdown of the "law of one price", leading to FVA, and CVA charges have led to the introduction of complex multi-factor models to jointly model the counterparty credit risk with the EPE/ENE of the portfolio. Before the financial crisis, "one-way risk" was a theoretical concern only, and institutions were happy to model CCR and exposures independently.The next things, in my opinion, are going to be:(i) the fallout from the concentration of CCPs. I think they're the new "too big to fail" and, despite risk models and prevention systems in place, nothing is invincible. (ii) "Liquidity smiles"
 
User avatar
Gamal
Posts: 2362
Joined: February 26th, 2004, 8:41 am

Is counterparty risk the new smile?

January 23rd, 2013, 4:31 pm

QuoteOriginally posted by: Liquidity smilesAnd reality bites.
 
User avatar
quartz
Posts: 427
Joined: June 28th, 2005, 12:33 pm

Is counterparty risk the new smile?

January 23rd, 2013, 5:48 pm

QuoteOriginally posted by: PvalAnal85Hi,I'm not so sure that Heston-parameter smile-fit modelling ever really went away. What we've seen is perhaps a vast unwind in the kind of complex structures (due to financial crisis) that made Heston calibration necessary in the first place. In addition, as models become "canonical", they are implemented at banks and hedge funds, and there is less of a need to discuss theoretical foundations of the model (as they're established in practice)....(i) the fallout from the concentration of CCPs. I think they're the new "too big to fail" and, despite risk models and prevention systems in place, nothing is invincible. (ii) "Liquidity smiles"Sure, Heston became a boring commodity, just like p-value in medicine, or oil/coal in energy: that didnt make it any better even though the unwind brought some exhogenous silencing :-)Btw, can we say that student t is the new gaussian copula?I agree on CCPs, regulators are shouting: "Hey, look here: if even these fail, they'd really fail big time.". To which the obvious answer is "If I can game these, what a game then!". A solution looking for another problem, and humans are ingenuous... err, I meant ingenious, of course! :-)
 
User avatar
Herd
Topic Author
Posts: 562
Joined: October 2nd, 2003, 12:48 pm

Is counterparty risk the new smile?

January 23rd, 2013, 8:28 pm

re t-copula and gaussian copula: gaussian copula is fine. it's just that one shouldnt use it with one single correlation.Historical estimation (unstationarity) goes wrong I guess because of new policies, national or international.The rules of the game change constantly, so it might be dangerous to use data from the past, when the laws were different.I suppose one can explain the credit boom that way (cheao easy credit, maybe equity release when house prices goes up was a new thing too?).And London/Paris house prices defy the rules because of a relatively new phenomenon: free movement of people/capital...And the euro was a new rule too...New rules create a new game... And it's tough seeing what are the new dangers.In finance (and economics) there are lots of new rules at the moment!
 
User avatar
frenchX
Posts: 5911
Joined: March 29th, 2010, 6:54 pm

Is counterparty risk the new smile?

January 23rd, 2013, 8:45 pm

Variance derivatives are becoming more and more vanilla so "smile models" are now a bit more complicated than original Heston one since bankers have to fit both the equity implied vol surface and the variance swap term structure. For a big challenge is risk decomposition: Your portfolio risk = Volatility risk+Fat tail risk+CCP risk+Interest rate risk+FX risk+etc... and each risk is modelled by a different model. The problem is that in finance risks are intimately intricated, correlated and shuffled in a very involved way and so additive decomposition of risks is inefficient and using a different model for each "orthogonal risk" is dangerous in my opinion especially at the portfolio or the bank level. The new smile would be a consistent smile analysis including credit, interest rate and FX risks in addition to the simpler stoch vol + jump models. Hybrid derivatives would help to hedge away those "risk intrications"
ABOUT WILMOTT

PW by JB

Wilmott.com has been "Serving the Quantitative Finance Community" since 2001. Continued...


Twitter LinkedIn Instagram

JOBS BOARD

JOBS BOARD

Looking for a quant job, risk, algo trading,...? Browse jobs here...


GZIP: On