• We are pleased to announce that the winner of our Feedback Prize Draw for the Winter 2024-25 session and winning £150 of gift vouchers is Zhao Liang Tay. Congratulations to Zhao Liang. If you fancy winning £150 worth of gift vouchers (from a major UK store) for the Summer 2025 exam sitting for just a few minutes of your time throughout the session, please see our website at https://www.acted.co.uk/further-info.html?pat=feedback#feedback-prize for more information on how you can make sure your name is included in the draw at the end of the session.
  • Please be advised that the SP1, SP5 and SP7 X1 deadline is the 14th July and not the 17th June as first stated. Please accept out apologies for any confusion caused.

Chapter 5.3 - Archimedean Copulas

V

Viki2010

Member
In the second table in this chapter, there is a formula listed which is TAU for each copula. I was wondering how is it derived?

I assume it is used for estimating alfa, after setting the formula equal to the emprisical value of TAU. Right?
 
In the second table in this chapter, there is a formula listed which is TAU for each copula. I was wondering how is it derived?

I assume it is used for estimating alfa, after setting the formula equal to the emprisical value of TAU. Right?

Yes, the formula is used for calibrating a copula,

Derived via integration (not in syllabus).
 
Last edited by a moderator:
Hello, going through the copula chapter again, I was wondering when can FRANK copula be applied in practice, given it does not have an upper or lower tail dependance?

The answer can be found in ASET April 2012 - Q 3 - thus please disregard this question.
 
Last edited by a moderator:
As Edwin states, there are integral formulae for tau & rho that lead to an expression in the copula parameter. It is beyond the syllabus but is something touched on in tutorials for the curious!
 
Hi Viki

I think that Sweeting says that the Frank copula has been used in joint life annuities in practice... maybe take a look in the textbook to confirm this.
 
Hi Viki

I think that Sweeting says that the Frank copula has been used in joint life annuities in practice... maybe take a look in the textbook to confirm this.

Thanks Alastair, I had a look earlier and did not find any applications of Frank copula. This type of copula is described in chapter 10.4.7 and it is the only one without any example.

I've found a note on its application in the exam answers - "modelling of equity indexes and bonds, which are not correlated and don't move together"
 
There are some scholarly articles on the Frank copula, for example this one looks at different copulas (inc Frank) to consider contagion risk (or lack thereof) between countries' markets. For example it concludes that Frank is the best approach for considering Thailand vs Indonesia markets in relatively tranquil periods.
 
Hello, I have not seen a lot of examples of applications of Fundamental copulas and Implicit copulas. Is that something we should know for the exam?

I don't think there are any examples of the applications in the Act Ed notes.
 
Hello, I have not seen a lot of examples of applications of Fundamental copulas and Implicit copulas. Is that something we should know for the exam?

I don't think there are any examples of the applications in the Act Ed notes.

An implicit copula example is the Gaussian copula and the student - t copula. Uses;-

Gaussian Copulas have zero tail dependence and are used in the pricing of Synthetic CDO with underlings made up of a basket of CDS. You should remember some Xi Li guy!

Student - t copula is used in a similar way for the valuation of synthetic CDO, Hull and White show that a good fit to the market is obtained when marginal distributions for F (a common factor affecting defaults for all companies) and Zi (a factor affecting only company i) follow Student t distributions with four degrees of freedom. They call this the double t copula.

The problem is the joint estimation of the covariance matrix and the degrees of freedom for the student - t, however you can use the degrees of freedom that results in a good fit to the tails in your data.

As for fundamental copulas, these aren't really copulas. They are special cases in which marginals can be "coupled" depending on the associated dependence...

...for example set rho to 100% for the Gaussian Copula (try bivariate case) and you will get the "minimum" = "comonotonic" = lower Frechet-Hoeffding bound COPULA! Set the correlation to 0 and you should get the independence copula.

Proof;-
C(u,v) = Phi_1 ( Phi^-1(u) , Phi^-1(v) ) = P ( X < Phi^-1(u) and X < Phi^-1(v) ) = Phi ( min ( Phi^-1(u), PHi^-1(v) ) ) = Phi ( Phi^-1( min ( u,v ) ) ) = min (u ,v)

ALSO see Sweeting page 303 it shows how x is a monotonic transformation of y i.e X=f(Y) for a minimum copula and gives the sterling example if you want to see an example in which the Frechet - Hoeffding copulas can OCCUR...

....the answer to your question is they are axioms that describe what is going on, not a tool for any use. But maybe Tutors can help here if I am wrong.
 
Last edited by a moderator:
Hello, going through the copula chapter again, I was wondering when can FRANK copula be applied in practice, given it does not have an upper or lower tail dependance?

The answer can be found in ASET April 2012 - Q 3 - thus please disregard this question.

Also the frank copula can be used in the Multivariate Structural model to evaluate the joint probability of insolvency of two firms, see Sweeting page 355-356.
 
Back
Top