MARGIN REQUIREMENT DETERMINATION AND MODELING FOR CLEARED CREDIT

• US 20150332404A1
• Filed: 05/07/2015
• Published: 11/19/2015
• Est. Priority Date: 05/16/2014
• Status: Active Grant

• Current Assignee: Chicago Mercantile Exchange, Inc. (CME Group Incorporated)
• Original Assignee: Chicago Mercantile Exchange, Inc. (CME Group Incorporated)
• Inventors: Wang, Chao, Baysal, Evren, Ding, Kailin, Li, Nick, Lu, Lu, Xythalis, Panos, Yang, Alice, Zhai, Jun
• Granted Patent: US 10,430,880 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06Q 40/04 Trading; Exchange, e.g. stocks, commodities, derivatives or currency exchange
• CPC Class Codes: G06Q 40/06 Asset management; Financial planning or analysis

Abstract

Systems and methods are provided for calculating margin requirements and stress testing exposures of cleared credit portfolios. These margin requirements are calculated using the following components: spread risk, idiosyncratic risk, interest rate, and liquidity risk. The calculation of these risk components is accomplished with a detailed statistical analysis of the risk factors underlying instruments, such as a credit default swap instrument.

Claims

1. A credit default swap (CDS) risk modeling computing system comprising:

at least one processor; and

one or more non-transitory memory devices communicatively coupled to the at least one processor, the non-transitory memory devices storing instructions that, when executed by the at least one processor, cause the CDS risk modeling computing system to;

calculate, using a spread risk factor calculator, a spread risk factor corresponding to a value at risk (VaR) associated with a plurality of correlation scenario sets, wherein the correlation scenario sets correspond to characteristics of at least one of a single name credit default swap or an index credit default swap of a portfolio;

calculate, using an idiosyncratic risk factor calculator, an idiosyncratic risk factor corresponding to a jump-to-default (JTD) charge and a jump-to-health (JTH) charge associated with the portfolio;

calculate, using an interest rate risk factor calculator, an interest rate risk factor corresponding to losses associated with the portfolio due to a change in interest rates;

calculate, using a liquidity risk factor calculator, a liquidity risk factor corresponding to a liquidity charge associated with the portfolio; and

calculate, by a margin calculator, a margin requirement for the portfolio based, at least in part on the spread risk factor, the idiosyncratic risk factor, the interest rate risk factor, and the liquidity risk factor.

2. The CDS risk modeling computing system of claim 1, wherein the one or more non-transitory memory devices further store instructions that, when executed by the at least one processor, cause the spread risk factor calculator to:

model, using a Monte Carlo-based scenario generator, a plurality of risk factors using a symmetric t-copula with four degrees of freedom, wherein an empirical copula associated with risk factor pairs is used to justify a choice of a multivariate copula distribution.

3. The CDS risk modeling computing system of claim 2, wherein the one or more non-transitory memory devices further store instructions that, when executed by the at least one processor, cause the scenario generator to

receive, at an input to the scenario generator, one or more correlation scenarios comprising a historical correlation matrix, a systemic correlation matrix, and a basis correlation matrix;

and

estimate one or more of a base value-at-risk (VaR), a basis VaR, and a systematic VaR based on the one or more correlation scenarios.

4. The CDS risk modeling computing system of claim 3, wherein the one or more non-transitory memory devices further store instructions that, when executed by the at least one processor, cause the scenario generator to

calculate a margin spread risk factor as a sum of the base VaR and a fraction of a maximum of either the basis VaR and the systematic VaR.

5. The CDS risk modeling computing system of claim 3, wherein the one or more non-transitory memory devices further store instructions that, when executed by the at least one processor, cause the scenario generator to

calculate a stress spread risk factor as a sum of the base VaR and a fraction of a maximum of either the basis VaR and the systematic VaR, wherein a first quantile of profit and loss (P&

L) distributions associated with a stress spread risk factor is greater than a second quantile of P&

L distributions associated with a margin spread risk factor.

6. The CDS risk modeling computing system of claim 2, wherein the one or more non-transitory memory devices further store instructions that, when executed by the at least one processor, cause the scenario generator to:

scale outputs from the symmetric t-copula by a corresponding marginal t-distribution and a forecasted exponential weighted moving average (EWMA) volatility.

7. The CDS risk modeling computing system of claim 6, wherein the one or more non-transitory memory devices further store instructions that, when executed by the at least one processor, cause the scenario generator to scale the EWMA volatility using a 1-day shock value and a margin period-of-risk shock value.

8. The CDS risk modeling computing system of claim 7, wherein a multiplier for a EWMA forecast is calibrated to a result of a cross sectional analysis of post or pre Lehman EWMA forecasts across different risk factors.

9. The CDS risk modeling computing system of claim 1, wherein the one or more non-transitory memory devices further store instructions that, when executed by the at least one processor, cause the idiosyncratic risk factor calculator to:

calculate an overall portfolio VaR associated with the portfolio;

calculate a JTD value-at-risk (VaR) associated with each single name position associated with the portfolio, wherein each JTD VaR comprises a default charge associated with a particular single name position and a remaining portfolio VaR corresponding to a remaining portion of the portfolio after removing the particular single name position;

calculate a maximum JTD VaR of the JTD VaR associated with each single name position; and

calculate the JTD charge as a difference between the maximum JTD VaR and the overall portfolio VaR.

10. The CDS risk modeling computing system of claim 9, wherein the one or more non-transitory memory devices further store instructions that, when executed by the at least one processor, cause the idiosyncratic risk factor calculator to:

calculate the remaining portfolio VaR associated with the portfolio after removing the particular single name position, wherein each index position in the remaining portion of the portfolio is adjusted to account for the removal of the particular single name position; and

calculate the default charge associated with the particular single name position as a difference between a current price of the particular single name position and a minimum recovery rate observed through a history associated with the particular single name position.

11. The CDS risk modeling computing system of claim 9, wherein, the one or more non-transitory memory devices further store instructions that, when executed by the at least one processor, cause the idiosyncratic risk factor calculator to:

calculate a margin JTD charge associated with the portfolio based on a historical correlation scenario set, wherein the margin JTD charge is used in calculating a margin requirement associated with the portfolio.

12. The CDS risk modeling computing system of claim 9, wherein, the one or more non-transitory memory devices further store instructions that, when executed by the at least one processor, cause the idiosyncratic risk factor calculator to:

calculate a historical JTD charge using a historical correlation scenario set;

calculate a basis JTD charge using a basis correlation scenario set;

calculate a systematic JTD charge using a systematic correlation scenario set; and

calculate a stress JTD charge associated with the portfolio as a maximum of the historical JTD charge, the basis JTD charge and the systematic JTD charge, wherein the stress JTD charge is used in determining a stress requirement associated with the portfolio.

13. The CDS risk modeling computing system of claim 1, wherein the one or more non-transitory memory devices further store instructions that, when executed by the at least one processor, cause the idiosyncratic risk factor calculator to:

calculate an overall portfolio VaR associated with the portfolio;

calculate a JTH value-at-risk (VaR) associated with each single name position associated with the portfolio, wherein each JTH VaR comprises a default charge associated with a particular single name position and a remaining portfolio VaR corresponding to a remaining portion of the portfolio after removing the particular single name position;

calculate a maximum JTH VaR of the JTH VaR associated with each single name position; and

calculate the JTH charge as a difference between the maximum JTH VaR and the overall portfolio VaR.

14. The CDS risk modeling computing system of claim 13, wherein, the one or more non-transitory memory devices further store instructions that, when executed by the at least one processor, cause the idiosyncratic risk factor calculator to:

calculate a margin JTH charge associated with the portfolio based on a historical correlation scenario set, wherein the margin JTH charge is used in calculating a margin requirement associated with the portfolio; and

calculate a stress JTH charge associated with the portfolio as a maximum of a historical JTD charge calculated using a historical correlation data set, a basis JTD charge calculated using a basis correlation data set and a systematic JTD charge calculated with a systematic correlation data set, wherein the stress JTD charge is used in determining a stress requirement associated with the portfolio.

15. The CDS risk modeling computing system of claim 1, wherein the one or more non-transitory memory devices further store instructions that, when executed by the at least one processor, cause the interest rate risk factor calculator to:

calculate an up-shock loss associated with an up shock to an interest rate curve used in CDS pricing;

calculate a down-shock loss associated with a down shock to the interest rate curve used in CDS pricing; and

calculate the interest rate risk factor as a maximum of the up-shock loss and the down-shock loss.

16. The CDS risk modeling computing system of claim 15, wherein a size of the up-shock and a size of the down shock are calibrated to a reference pivot rate.

17. The CDS risk modeling computing system of claim 1, wherein the one or more non-transitory memory devices further store instructions that, when executed by the at least one processor, cause the liquidity risk factor calculator to:

calculate an outright exposure to an investment grade (IG) sub-portfolio of the portfolio;

calculate an outright exposure to a high yield (HY) sub-portfolio of the portfolio;

calculate a basis exposure to at least one of an index-based CDS sub-portfolio and a basis exposure to a single name CDS sub-portfolio of the portfolio; and

calculate the liquidity risk factor based on the outright exposure of the IG sub-portfolio, the outright exposure of the HY sub-portfolio and the basis exposure of the CDS portfolio.

18. An apparatus comprising:

at least one processor; and

one or more non-transitory memory devices communicatively coupled to the at least one processor, the non-transitory memory devices storing instructions that, when executed by the at least one processor, cause the apparatus to;

calculate, using a spread risk factor calculator, a spread risk factor corresponding to a value at risk (VaR) associated with a plurality of correlation scenario sets, wherein the correlation scenario sets correspond to characteristics of at least one of a single name credit default swap or an index credit default swap of a portfolio;

calculate, using an idiosyncratic risk factor calculator, an idiosyncratic risk factor corresponding to a jump-to-default (JTD) charge and a jump-to-health (JTH) charge associated with the portfolio;

calculate, using an interest rate risk factor calculator, an interest rate risk factor corresponding to losses associated with the portfolio due to a change in interest rates; and

calculate, using a liquidity risk factor calculator, a liquidity risk factor corresponding to a liquidity charge associated with the portfolio.

19. The apparatus of claim 18, wherein the instructions, when executed by the at least one processor, further cause the apparatus to:

calculate, by a margin calculator, a margin requirement for the portfolio based, at least in part on the spread risk factor, the idiosyncratic risk factor, the interest rate risk factor, and the liquidity risk factor; and

calculate, by a stress requirement calculator, a stress requirement for the portfolio based, at least in part on the spread risk factor, the idiosyncratic risk factor, the interest rate risk factor, and the liquidity risk factor.

20. A method comprising:

calculating, by a spread risk calculator using a Monte Carlo-based scenario generator, a spread risk factor corresponding to a value at risk (VaR) associated with a plurality of correlation scenario sets, wherein the correlation scenario sets correspond to characteristics of at least one of a single name credit default swap or an index credit default swap of a portfolio and wherein the Monte Carlo-based scenario generator uses a symmetric t-copula with four degrees of freedom;

calculating, using an idiosyncratic risk factor calculator, an idiosyncratic risk factor corresponding to a portfolio jump-to-default (JTD) charge and a portfolio jump-to-health (JTH) charge associated with the portfolio, wherein the portfolio JTD charge corresponds to a maximum of a plurality of single name JTD charges associated with the portfolio and the portfolio JTH charge corresponds to a maximum of a plurality of single name JTH charges associated with the portfolio;

calculating, using an interest rate risk factor calculator, an interest rate risk factor corresponding to losses associated with the portfolio due to a change in interest rates, wherein the interest rate risk factor comprises a maximum of an up shock loss and a down shock loss associated with an interest rate curve used in pricing a CDS holding associated with the portfolio;

calculating, using a liquidity risk factor calculator, a liquidity risk factor corresponding to a liquidity charge associated with the portfolio, wherein the liquidity risk factor is based on an outright exposure of an investment grade sub-portfolio of the portfolio, an outright exposure of a high yield sub-portfolio of the portfolio and a basis exposure of the portfolio;

calculating, by a margin calculator, a margin requirement for the portfolio based, at least in part on the spread risk factor, the idiosyncratic risk factor, the interest rate risk factor, and the liquidity risk factor; and

calculating, by a stress requirement calculator, a stress requirement for the portfolio based, at least in part on the spread risk factor, the idiosyncratic risk factor, the interest rate risk factor, and the liquidity risk factor.

21. The method of claim 20, further comprising:

receiving, via a user interface, a request to calculate one of a margin requirement for a portfolio of credit default swaps or a stress evaluation of the portfolio; and

based on the request received at the user interface, associating the margin requirement to the portfolio by a clearinghouse computing system or communicate the stress requirement to the clearinghouse computing system for inclusion in a risk management procedure by a risk management module of the clearinghouse computing system.