I was seeking for good literature about CVaR and i have to say, that this lecture helped me alot understand better how do calculate the CVaR. Thank you very much.
I agree with Minh Bui, the formula that used to calculate capital requirement in Excel is not the same it shows on the K formula. The formula that is the same as the box should be =EAD*LGD*((NORMSDIST((NORMSINV(PD)+rho^0.5*NORMSINV(CL))/(1-rho)^0.5))-PD)*MaturityAdj. But they have the same result. Thanks so much for the lecture, it is so helpful to understand this subject!
I doubt it (as the excel XLS has survived several FRM seasons). More likely i think is simply that the formula itself is a non-trivial copula. But if you want to share the specific discrepancy, i can try to address it, thanks,
Hello David. Thank you for the video. Very helpful. I made some calculations which are almost the same as yours. I took your 18% (ro) to have the same hypothesis as yours. Even if the calculation of this correlation factor is to be calculated and if a bit different (I took the formulas from the BIS website). Banque des règlements internationaux (BRI) in french. My point is that the K (capital requirement) I found K is 9000 lower than yours. 9000 Which is the EL (Expected Loss). I do not see why we have this difference. According to the formula of K, the result should be $128k. Not 137… Can you please help me to understand ?
same, i also calculate the K by using the BIS formula and I also find the difference I wonder whether you figured out whether this difference happened?
UL - EL isn't meaningful. Capital covers UL (and presumes EL is covered as an expense; ie, in product pricing). "Absolute CVaR" = UL + EL, and CVaR (aka, relative VaR) = UL.
I have a trouble of seeing the similarity between the function you wrote and the excel calculation when you highlighted it. Is that a mistake of calculation from you?
actually I was wrong about what I said above. btw I still think you are not right about the Maturity Adjustment and for people, b(PD)=[ (0.11852-0.05478*ln(PD) ]^2 I think for the Maturity Adjustment, the top should be 1+(M-2.5)*b(PD)
Wzorki do excela dla polaków ;):Unexpectedloss =((B3*((ROZKŁAD.NORMALNY.S(((1/(1-B8))^0,5*ROZKŁAD.NORMALNY.S.ODW(B2)+(B8/(1-B8))^0,5*ROZKŁAD.NORMALNY.S.ODW(B10)))))-B2*B3)*B11)*B4, Maturity =(1+(B5-2,5)*((0,11852-0,05478*LN(B2))^2))/(1-1,5*((0,11852-0,05478*LN(B2))^2)) Jeżeli pd-B2,lgd-B3,ead-B4,m-B5, itp. Jeżeli potraficie to tak skrócić jak na filmiku to piszcie ;)
You are the best Risk management teacher I have ever encountered ! I owe you several exam passes
Thank you so much for your kind words! I really appreciate your support.
I was seeking for good literature about CVaR and i have to say, that this lecture helped me alot understand better how do calculate the CVaR. Thank you very much.
I agree with Minh Bui, the formula that used to calculate capital requirement in Excel is not the same it shows on the K formula. The formula that is the same as the box should be =EAD*LGD*((NORMSDIST((NORMSINV(PD)+rho^0.5*NORMSINV(CL))/(1-rho)^0.5))-PD)*MaturityAdj. But they have the same result. Thanks so much for the lecture, it is so helpful to understand this subject!
Good Afternoon Mr Harper,
Thx for the content, why in the excel cell you did not multiply the PD by LGD ? You only put - PD why ?
In advance thx
Yes. I agree with the former comment. The maturity adjustment formula is a bit puzzling. What does b mean there?
Ok, once capital is calculated and set aside. In what form the capital will be held, is it in cash form or any other highly liquid assets?
What u explained so far was quite intuitive except for maturity adjustment. Could u explain more?
I doubt it (as the excel XLS has survived several FRM seasons). More likely i think is simply that the formula itself is a non-trivial copula. But if you want to share the specific discrepancy, i can try to address it, thanks,
Hello David.
Thank you for the video. Very helpful.
I made some calculations which are almost the same as yours.
I took your 18% (ro) to have the same hypothesis as yours.
Even if the calculation of this correlation factor is to be calculated and if a bit different (I took the formulas from the BIS website). Banque des règlements internationaux (BRI) in french.
My point is that the K (capital requirement)
I found K is 9000 lower than yours.
9000 Which is the EL (Expected Loss).
I do not see why we have this difference.
According to the formula of K, the result should be $128k. Not 137…
Can you please help me to understand ?
same, i also calculate the K by using the BIS formula and I also find the difference
I wonder whether you figured out whether this difference happened?
Thanks, I had missed this one!
Excellent Explanation
Why cap req added UL to EL. It should be UL-EL right?
UL - EL isn't meaningful. Capital covers UL (and presumes EL is covered as an expense; ie, in product pricing). "Absolute CVaR" = UL + EL, and CVaR (aka, relative VaR) = UL.
I have a trouble of seeing the similarity between the function you wrote and the excel calculation when you highlighted it. Is that a mistake of calculation from you?
actually I was wrong about what I said above. btw I still think you are not right about the Maturity Adjustment
and for people, b(PD)=[ (0.11852-0.05478*ln(PD) ]^2
I think for the Maturity Adjustment, the top should be 1+(M-2.5)*b(PD)
dksieunhan: I think you are right, but in the excel formula the pattern is as You wrote;) I just checked it seems ok.
Thanks man!
can you see my second comment below, I think it is the only problem. thank you!
Wzorki do excela dla polaków ;):Unexpectedloss =((B3*((ROZKŁAD.NORMALNY.S(((1/(1-B8))^0,5*ROZKŁAD.NORMALNY.S.ODW(B2)+(B8/(1-B8))^0,5*ROZKŁAD.NORMALNY.S.ODW(B10)))))-B2*B3)*B11)*B4, Maturity =(1+(B5-2,5)*((0,11852-0,05478*LN(B2))^2))/(1-1,5*((0,11852-0,05478*LN(B2))^2)) Jeżeli pd-B2,lgd-B3,ead-B4,m-B5, itp. Jeżeli potraficie to tak skrócić jak na filmiku to piszcie ;)