Futures margin modeling system

• US 10,643,278 B2
• Filed: 01/20/2016
• Issued: 05/05/2020
• Est. Priority Date: 01/20/2016
• Status: Active Grant

• Current Assignee: Chicago Mercantile Exchange, Inc. (CME Group Incorporated)
• Original Assignee: Chicago Mercantile Exchange, Inc. (CME Group Incorporated)
• Inventors: Weng, Jennifer, Joshi, Nikhil, Chen, Guo, Yang, Siwen, Yang, Zijiang
• Primary Examiner(s): Sigmond, Bennett M
• Assistant Examiner(s): Cunningham, II, Gregory S
• Application Number:
  US15/001,997
• Publication Number:
  US 20170206601A1
• Time in Patent Office: 1,567 Days
• Field of Search: None
• CPC Class Codes: G06Q 40/04 Trading; Exchange, e.g. stocks, commodities, derivatives or currency exchange

Abstract

A clearinghouse computing device may be configured to generate a margin requirement for a portfolio of financial products and may include a processor to process instructions that cause the clearinghouse computing device to retrieve a plurality of pricing records from a historical pricing database, process the plurality of pricing records to generate rolling time series pricing records for at least one financial product having a plurality of dimensions, reduce the number of dimensions from a starting dimension to a reduced dimension, perform variance scaling and correlation scaling on the reduced dimension rolling time series pricing records, and generate a margin requirement based on a value-at-risk calculation.

Claims

1. A clearinghouse computing device configured to generate a margin requirement for a portfolio of financial products, comprising:

a processor; and

a non-transitory memory device storing instructions that, when executed by the processor, cause the clearinghouse computing device to;

(a) generate, by a time series generator of a clearinghouse computer system, continuous time series of pricing information corresponding to a financial product held in a portfolio, the continuous time series of pricing information comprising a plurality of dimensions each corresponding to a factor contributing to dynamics of the continuous time series;

(b) calculate, by a dimension reduction module of the clearinghouse computer system, a dimension-reduced projection of the continuous time series of pricing information based on a principle component analysis (PCA) technique the dimension reduced projection using less than all of the factors to represent the dynamics of the continuous time series;

(c) calculate, by a variance scaling module of the clearinghouse computer system, a volatility normalization of the dimension reduced projection by performing volatility normalization of scores and volatility normalization of error terms of the dimension reduced projection of the continuous time series of pricing information to produce a plurality of curves;

(d) determine, by a covariance scaling module of the clearinghouse computer system, an inter-curve and intra-curve correlation between the plurality of curves; and

(e) generate, by a value at risk estimation module of the clearinghouse computer system, an estimated value at risk based on the inter-curve and intra-curve correlation between the plurality of curves.

2. The clearinghouse computing device of claim 1 wherein a calculated margin requirement corresponds to the value at risk generated by the value at risk estimation module.

3. The clearinghouse computing device of claim 1, wherein the portfolio comprises two or more financial products, wherein the non-transitory memory device comprises instructions that, when executed by the processor, cause the clearinghouse computing device to

perform steps (a)-(e) for each of two or more financial products in the portfolio;

and

calculate, by a margin calculator, a portfolio margin requirement based on the value-at-risk of for each of the two or more financial products.

4. The clearinghouse computing device of claim 1, wherein instructions cause the clearinghouse computing device, as part of step (a), to extrapolate, by the time series generator, one or more missing data points.

5. The clearinghouse computing device of claim 1, wherein the missing data points are at the tails of a pricing curve of the financial product.

6. The clearinghouse computing device of claim 1, wherein the dimension reduced projection comprises a number of dimensions reduced from a beginning number of dimensions by at least 80 percent.

7. The clearinghouse computing device of claim 1, wherein the dimension reduced projection comprises a number of dimensions reduced from a beginning number of dimensions to less than 10 dimensions.

8. The clearinghouse computing device of claim 1, wherein (c) is performed using a dual lambda approach wherein a volatility of the scores is set to the maximum of a fast lambda EWMA and a slow lambda EWMA.

9. The clearinghouse computing device of claim 1, further comprising a user interface device including a display device and a user input device, wherein the non-transitory memory device stores instructions that, when executed by the processor, cause the clearinghouse computing device to:

present, to the user via the display device, at least one backtesting screen comprising a result from a test; and

receive, via the user input device, an indication whether a model for determining a portfolio margin requirement has passed a qualitative validation test.

10. A clearinghouse computing system configured to generate a margin requirement for a portfolio of financial products, comprising:

a historical pricing database storing pricing information for a plurality of financial products;

a clearinghouse computing device communicatively coupled to the historical pricing database and comprising;

a processor; and

a non-transitory memory device storing instructions that, when executed by the processor, cause the clearinghouse computing device to;

(a) generate, by a time series generator of a clearinghouse computer system, continuous time series of pricing information corresponding to a financial product held in a portfolio, the continuous time series of pricing information comprising a plurality of dimensions each corresponding to a factor contributing to dynamics of the continuous time series;

(b) calculate, by a dimension reduction module of the clearinghouse computer system, a dimension-reduced projection of the continuous time series of pricing information based on a principle component analysis (PCA) technique the dimension reduced projection using less than all of the factors to represent the dynamics of the continuous time series;

(c) calculate, by a variance scaling module of the clearinghouse computer system, a volatility normalization of the dimension reduced projection by performing volatility normalization of scores and volatility normalization of error terms of the dimension reduced projection of the continuous time series of pricing information to produce a plurality of curves;

(d) determine, by a covariance scaling module of the clearinghouse computer system, an inter-curve and intra-curve correlation between the plurality of curves; and

(e) generate, by a value at risk estimation module of the clearinghouse computer system, an estimated value at risk based on the inter-curve and intra-curve correlation between the plurality of curves.

11. The clearinghouse computing device of claim 10 wherein a calculated margin requirement corresponds to the value at risk generated by the value at risk estimation module.

12. The clearinghouse computing device of claim 10, wherein the portfolio comprises two or more financial products, wherein the non-transitory memory device comprises instructions that, when executed by the processor, cause the clearinghouse computing device to

perform steps (a)-(e) for each of two or more financial products in the portfolio;

and

calculate, by a margin calculator, a portfolio margin requirement based on the value-at-risk of for each of the two or more financial products.

13. The clearinghouse computing device of claim 10, wherein instructions cause the clearinghouse computing device, as part of step (a), to extrapolate, by the time series generator, one or more missing data points.

14. The clearinghouse computing device of claim 10, wherein the missing data points are at the tails of a pricing curve of the financial product.

15. The clearinghouse computing device of claim 10, wherein the dimension reduced projection comprises a number of dimensions reduced from a beginning number of dimensions by at least 80 percent.

16. The clearinghouse computing device of claim 10, wherein the dimension reduced projection comprises a number of dimensions reduced from a beginning number of dimensions to less than 10 dimensions.

17. The clearinghouse computing device of claim 10, wherein (c) is performed using a dual lambda approach wherein a volatility of the scores is set to the maximum of a fast lambda EWMA and a slow lambda EWMA.

18. The clearinghouse computing device of claim 10, further comprising a user interface device including a display device and a user input device, wherein the non-transitory memory device stores instructions that, when executed by the processor, cause the clearinghouse computing device to:

present, to the user via the display device, at least one backtesting screen comprising a result from a test; and

receive, via the user input device, an indication whether a model for determining a portfolio margin requirement has passed a qualitative validation test.

19. A method performed by a clearinghouse computing device, comprising:

(a) generating, by a time series generator of the clearinghouse computer device, continuous time series of pricing information corresponding to a financial product held in a portfolio, the continuous time series of pricing information comprising a plurality of dimensions each corresponding to a factor contributing to dynamics of the continuous time series;

(b) calculating, by a dimension reduction module of the clearinghouse computer device, a dimension-reduced projection of the continuous time series of pricing information based on a principle component analysis (PCA) technique the dimension reduced projection using less than all of the factors to represent the dynamics of the continuous time series;

(c) scaling, by a variance scaling module of the clearinghouse computer device, by performing volatility normalization of scores and volatility normalization of error terms of the dimension reduced projection of the continuous time series of pricing information to produce a plurality of curves;

(d) determining, by a covariance scaling module of the clearinghouse computer device, an inter-curve and intra-curve correlation between the plurality of curves; and

(e) generating, by a value at risk estimation module of the clearinghouse computer device, an estimated value at risk based on the inter-curve and intra-curve correlation between the plurality of curves.

20. The method of claim 19, comprising:

performing steps (a)-(e) for each of two or more financial products in the portfolio; and

calculating, by a margin calculator, a portfolio margin requirement based on the value-at-risk of for each of the two or more financial products.

21. The method of claim 19, comprising:

validating, by the margin calculator of the clearinghouse computing device, a margin generation model using one or more of a qualitative backtesting method or a quantitative backtesting method.