System and method for determining a liquidity-adjusted value at risk (LA-VaR)

• US 8,234,201 B1
• Filed: 03/20/2009
• Issued: 07/31/2012
• Est. Priority Date: 08/01/2008
• Status: Active Grant

• Current Assignee: Morgan Stanley Services Group Inc. (Morgan Stanley)
• Original Assignee: Morgan Stanley
• Inventors: Canabarro, Eduardo
• Primary Examiner(s): Kazimi, Hani M
• Assistant Examiner(s): Preston, John
• Application Number:
  US12/408,240
• Time in Patent Office: 1,229 Days
• Field of Search: 705/36.R, 705/35
• US Class Current: 705/36.R
• CPC Class Codes: G06Q 40/06 Asset management; Financial planning or analysis

Abstract

Computer-based systems and methods for calculating a liquidity-adjusted VaR for a portfolio. The liquidity-adjusted VaR accounts for the different liquidities of the risks. The process for calculating the liquidity-adjusted VaR may include adding to the standard 1-day VaR only the losses produced by illiquid risks that are orthogonal to the space of liquid risks.

Claims

1. A computer-implemented method for calculating an x_n-day liquidity-adjusted value-at-risk (LA-VaR) measure for a financial portfolio, the method comprising:

determining, by a computer system, from an initial set of market risk factors for the portfolio, a core set of liquid risk factors for the portfolio, wherein the core set of liquid risk factors represents a space of liquid risks for the portfolio;

determining, by the computer system, a subset of illiquid risk factors from the initial set of risk factors via a linear regression of each illiquid risk factor on the space of liquid risks as represented by the core set of liquid risk factors;

storing, by the computer system, a liquidity time horizon for each illiquid risk factor;

calculating by the computer system a profit and loss (P&

L) distribution for the portfolio for a time interval [0,1 day];

calculating by the computer system one or more P&

L distributions for the portfolio for the time intervals day], wherein x_i>

1 day by simulating only the illiquid risk factors having a liquidity time horizon greater than or equal to x_n-days, where x_i-1=1 day, where i=1, . . . n, and where n≧

1;

combining by the computer system the P&

L distributions for the [0,1] time interval and the one or more (x_i-1,x_i day] time intervals; and

calculating by the computer system the x_n-day VaR for the portfolio based on the combined P&

L distributions,

wherein the computer system comprises at least one computer device, wherein the computer device comprises at least one processor circuit and at least one memory circuit.

2. The method of claim 1, wherein the step of determining the core set of liquid risk factors for the portfolio comprises using principal component analysis to determine the core set of liquid risks.

3. The method of claim 1, wherein the step of combining the P&

L distributions for the [0,1 day] time interval and the one or more (x_i-1,x_i day] time intervals comprises convolving the P&

L distributions for the [0,1 day] time interval and the one or more (x_i-1,x_i day] time intervals.

4. The method of claim 1, wherein:

the step of calculating the one or more P&

L distributions for the portfolio comprises

calculating P&

L distributions for the time intervals (1,5 days], (5,10 days], and (10,20 days];

the step of combining the P&

L distributions for the [0,1 day] time interval and the one or more (x_i-1,x_i day] time intervals comprises convolving the P&

L distributions for the [0,1 day] time interval, the (1, 5 day] time interval, the (5, 10 day] time interval, and the (10, 20 day] time interval; and

the step of calculating the x_n-day LA-VaR comprises calculating the 5-day LA-VaR, the 10-day LA-VaR, and the 20-day LA-VaR based on the convolved P&

L distributions for the [0,1 day] time interval, the (1, 5 day] time interval, the (5, 10 day] time interval, and the (10, 20 day] time interval.

5. The method of claim 4, wherein the step of calculating the one or more P&

L distributions for the portfolio comprises scaling the illiquid risk factors by a multiplicative factor to reflect the length of the time interval.

6. The method of claim 5, wherein the step of scaling the illiquid risk factors by a multiplicative factor comprises:

scaling a illiquid risk factor having liquidity time horizon equal to or greater than five days by √

{square root over (2)} when calculating the P&

L distribution for the (1, 5 day] time interval;

scaling a illiquid risk factor having liquidity time horizon equal to or greater than five days by √

{square root over (5/2)} when calculating the P&

L distribution for the (5, 10 day] time interval; and

scaling a illiquid risk factor having liquidity time horizon equal to or greater than five days by √

{square root over (5)} when calculating the P&

L distribution for the (10, 20 day] time interval.

7. The method of claim 1, wherein the step of determining the subset of illiquid risk factors from the initial set of risk factors via a linear regression of each illiquid risk factor on the space of liquid risks as represented by the core set of liquid risk factors comprises modeling the linear regression based on the relation:

{right arrow over (r)}_i={right arrow over (L)}·

{right arrow over (β

)}_i+{right arrow over (e)}_i

where;

{right arrow over (r)}_i is a (T×

1) vector of daily changes on the non-liquid risk factor i and T is a number of days in a time series;

{right arrow over (L)} is a matrix of dimensions (T×

m) containing the m principal components of the space of liquid risks;

{right arrow over (β

)}_i is the (m×

1) vector of sensitivities of risk factor {right arrow over (r)}_i to each component of the core set of liquid risk factors; and

{right arrow over (e)}_i is the (T×

1) matrix of residuals of factor {right arrow over (r)}_i and is orthogonal to the space of liquid risks.

8. A system for calculating an x_n-day liquidity-adjusted value-at-risk (LA-VaR) measure for a financial portfolio, the system comprising:

a computer data store; and

a computer device in communication with the computer data store, wherein the computer device comprises at least one processor circuit and at least one memory circuit, and wherein the computer device is programmed to;

determine, from an initial set of market risk factors for the portfolio, a core set of liquid risk factors for the portfolio, wherein the core set of liquid risk factors represents a space of liquid risks for the portfolio;

determine a subset of illiquid risk factors from the initial set of risk factors via a linear regression of each illiquid risk factor on the space of liquid risks as represented by the core set of liquid risk factors, wherein a liquidity time horizon for each illiquid risk factor is stored in the computer data store;

calculate a profit and loss (P&

L) distribution for the portfolio for a time interval [0,1 day];

calculate one or more P&

L distributions for the portfolio for the time intervals (x_i-1,x_i day], wherein x_i>

1 day by simulating only the illiquid risk factors having a liquidity time horizon greater than or equal to x, days, where x_i-1=1 day, where i=1, . . . n, and where n≧

1;

combine the P&

L distributions for the [0,1 day] time interval and the one or more (x_i-1,x_iday] time intervals; and

calculate the x_n-day VaR for the portfolio based on the combined P&

L distributions.

9. The system of claim 8, wherein the computer device is programmed to determine the core set of liquid risk factors for the portfolio using principal component analysis.

10. The system of claim 8, wherein the computer device is programmed to combine the P&

L distributions for the [0,1 day] time interval and the one or more (x_i-1,x_iday] time intervals by convolving the P&

L distributions for the [0,1 day] time interval and the one or more (x_i-1,x_iday] time intervals.

11. The system of claim 8, wherein the computer device is programmed to:

calculate the one or more P&

L distributions for the portfolio by calculating P&

L distributions for the time intervals (1,5 days], (5,10 days], and (10,20 days];

combine the P&

L distributions for the [0,1 day] time interval and the one or more (x_i-1,x_iday] time intervals by convolving the P&

L distributions for the [0,1 day] time interval, the (1, 5 day] time interval, the (5, 10 day] time interval, and the (10, 20 day] time interval; and

calculate the x_n-day LA-VaR by calculating the 5-day LA-VaR, the 10-day LA-VaR, and the 20-day LA-VaR based on the convolved P&

L distributions for the [0,1 day] time interval, the (1, 5 day] time interval, the (5, 10 day] time interval, and the (10, 20 day] time interval.

12. The system of claim 11, wherein the computer device is programmed to calculate the one or more P&

L distributions for the portfolio by scaling the illiquid risk factors by a multiplicative factor to reflect the length of the time interval.

13. The system of claim 12, wherein the computer device is programmed to scale the illiquid risk factors by a multiplicative factor by:

scaling a illiquid risk factor having liquidity time horizon equal to or greater than five days by √

{square root over (2)} when calculating the P&

L distribution for the (1, 5 day] time interval;

scaling a illiquid risk factor having liquidity time horizon equal to or greater than five days by √

{square root over (5/2)} when calculating the P&

L distribution for the (5, 10 day] time interval; and

scaling a illiquid risk factor having liquidity time horizon equal to or greater than five days by √

{square root over (5)} when calculating the P&

L distribution for the (10, 20 day] time interval.

14. The system of claim 8, wherein the computer device is programmed to determine the subset of illiquid risk factors from the initial set of risk factors via a linear regression of each illiquid risk factor on the space of liquid risks as represented by the core set of liquid risk factors by modeling the linear regression based on the relation:

{right arrow over (r)}_i={right arrow over (L)}·

{right arrow over (β

)}_i+{right arrow over (e)}_i

where;

{right arrow over (r)}_i is a (T×

1) vector of daily changes on the non-liquid risk factor i and T is a number of days in a time series;

{right arrow over (L)} is a matrix of dimensions (T×

m) containing the m principal components of the space of liquid risks;

{right arrow over (β

)}_i is the (m×

1) vector of sensitivities of risk factor {right arrow over (r)}_i to each component of the core set of liquid risk factors; and

{right arrow over (e)}_i is the (T×

1) matrix of residuals of factor {right arrow over (r)}_i and is orthogonal to the space of liquid risks.

15. A non-transitory storage medium having stored thereon instructions that when executed by a processor circuit cause the processor circuit to calculate an x_n-day liquidity-adjusted value-at-risk (LA-VaR) measure for a financial portfolio by:

determining, from an initial set of market risk factors for the portfolio, a core set of liquid risk factors for the portfolio, wherein the core set of liquid risk factors represents a space of liquid risks for the portfolio;

determining a subset of illiquid risk factors from the initial set of risk factors via a linear regression of each illiquid risk factor on the space of liquid risks as represented by the core set of liquid risk factors;

calculating a profit and loss (P&

L) distribution for the portfolio for a time interval [0,1 day];

calculating one or more P&

L distributions for the portfolio for the time intervals (x_i-1,x_iday], wherein x_i>

1 day by simulating only the illiquid risk factors having a liquidity time horizon greater than or equal to x_idays, where x_i-1=1 day, where i=1, . . . n, and where n≧

1;

combining the P&

L distributions for the [0,1 day] time interval and the one or more (x_i-1,x_iday] time intervals; and

calculating the x_n-day VaR for the portfolio based on the combined P&

L distributions.

16. The non-transitory storage medium of claim 15, wherein the computer readable medium has instructions stored thereon which when executed by the processor circuit causes the processor circuit to determine the core set of liquid risk factors for the portfolio using principal component analysis.

17. The non-transitory storage medium of claim 15, wherein the computer readable medium has instructions stored thereon which when executed by the processor circuit causes the processor circuit to combine the P&

L distributions for the [0,1 day] time interval and the one or more (x_i-1,x_i] time intervals by convolving the P&

L distributions for the [0,1 day] time interval and the one or more (x_i-1,x_i] time intervals.

18. The non-transitory storage medium of claim 15, wherein the computer readable medium has instructions stored thereon which when executed by the processor circuit causes the processor circuit to:

calculate the one or more P&

L distributions for the portfolio by calculating P&

L distributions for the time intervals (1,5 days], (5,10 days], and (10,20 days];

combine the P&

L distributions for the [0,1 day] time interval and the one or more (x_i-1,x_iday] time intervals by convolving the P&

L distributions for the [0,1 day] time interval, the (1, 5 day] time interval, the (5, 10 day] time interval, and the (10, 20 day] time interval; and

calculate the x_n-day LA-VaR by calculating the 5-day LA-VaR, the 10-day LA-VaR, and the 20-day LA-VaR based on the convolved P&

L distributions for the [0,1 day] time interval, the (1, 5 day] time interval, the (5, 10 day] time interval, and the (10, 20 day] time interval.

19. The non-transitory storage medium of claim 18, wherein the computer readable medium has instructions stored thereon which when executed by the processor circuit causes the processor circuit to calculate the one or more P&

L distributions for the portfolio by scaling the illiquid risk factors by a multiplicative factor to reflect the length of the time interval.

20. The non-transitory storage medium of claim 19, wherein the computer readable medium has instructions stored thereon which when executed by the processor circuit causes the processor circuit to scale the illiquid risk factors by a multiplicative factor by:

scaling a illiquid risk factor having liquidity time horizon equal to or greater than five days by √

{square root over (2)} when calculating the P&

L distribution for the (1, 5 day] time interval;

scaling a illiquid risk factor having liquidity time horizon equal to or greater than five days by √

{square root over (5/2)} when calculating the P&

L distribution for the (5, 10 day] time interval; and

scaling a illiquid risk factor having liquidity time horizon equal to or greater than five days by √

{square root over (5)} when calculating the P&

L distribution for the (10, 20 day] time interval.

21. The non-transitory storage medium of claim 15, wherein the computer readable medium has instructions stored thereon which when executed by the processor circuit causes the processor circuit to determine the subset of illiquid risk factors from the initial set of risk factors via a linear regression of each illiquid risk factor on the space of liquid risks as represented by the core set of liquid risk factors by modeling the linear regression based on the relation:

{right arrow over (r)}_i={right arrow over (L)}·

{right arrow over (β

)}_i+{right arrow over (e)}_i

where;

{right arrow over (r)}_i is a (T×

1) vector of daily changes on the non-liquid risk factor i and T is a number of days in a time series;

{right arrow over (L)} is a matrix of dimensions (T×

m) containing the m principal components of the space of liquid risks;

{right arrow over (β

)}_i is the (m×

1) vector of sensitivities of risk factor {right arrow over (r)}_i to each component of the core set of liquid risk factors; and

{right arrow over (e)}_i is the (T×

1) matrix of residuals of factor {right arrow over (r)}_i and is orthogonal to the space of liquid risks.

22. A computer-implemented method for calculating a liquidity-adjusted value-at-risk (LA-VaR) measure for a financial portfolio for a n-day VaR calculation period, the method comprising:

where the n-day VaR calculation period has a number m of sub-time intervals indexed by k=1, . . . , m, each sub-time interval k having a start time t_k-1 and an end time t_k, and t₀=0,

(a) determining, by a computer system, a core set of liquid risk factors available in a first sub-time interval [t₀, t₁], wherein the liquid risk factors available in the first sub-time interval are risk factors that have liquidity horizons less than or equal to the end time t₁ of the first sub-time interval and represent a space of liquid risks for the portfolio for the first sub-time interval, wherein risk factors with liquidity horizons greater than the end time t₁ of the first sub-time interval are illiquid risk factors in the first sub-time interval;

(b) extracting, by the computer system, liquid risk components of the illiquid risk factors in the first sub-time interval [t₀, t₁] via linear regression of each illiquid risk factor on the core set of liquid risk factor in the first sub-time interval, wherein residuals of the linear regression represent the illiquid risk factors in the first sub-time interval after extraction of the liquid risk components of the illiquid risk factors;

(c) calculating, by the computer system, a profit and loss (P&

L) distribution for the portfolio for the first sub-time interval [t₀, t₁] based on the residuals and the liquid risk factors for the first sub-time interval;

(d) repeating steps (a)-(c) for sub-time intervals indexed by k=2, . . . , m;

(e) computing, by the computer system, a combined P&

L distribution for the portfolio by combining the P&

L distributions for each of the sub-time intervals indexed by k=1, . . . , m; and

(f) calculating, by the computer system, the VaR for the portfolio for the n-day VaR calculation period based on the combined P&

L distribution.

23. The method of claim 22, wherein calculating the combined P&

L distribution comprises convolving the P&

L distributions calculated for k=1, . . . , m.

24. A system for calculating a liquidity-adjusted value-at-risk (LA-VaR) measure for a financial portfolio for a n-day VaR calculation period, the system comprising:

at least one computer device that comprises at least one processor circuit and at least one memory circuit, and wherein the at least one computer device is programmed to;

where the n-day VaR calculation period has a number m of sub-time intervals indexed by k=1, . . . , m, each sub-time interval k having a start time t_k-1, and an end time t_k, and t₀=0,

(a) determine a core set of liquid risk factors available in a first sub-time interval [t₀, t₁], wherein the liquid risk factors available in the first sub-time interval are risk factors that have liquidity horizons less than or equal to the end time t₁ of the first sub-time interval and represent a space of liquid risks for the portfolio for the first sub-time interval, wherein risk factors with liquidity horizons greater than the end time t₁ of the first sub-time interval are illiquid risk factors in the first sub-time interval;

(b) extract liquid risk components of the illiquid risk factors in the first sub-time interval [t₀, t₁] via linear regression of each illiquid risk factor on the core set of liquid risk factor in the first sub-time interval, wherein residuals of the linear regression represent the illiquid risk factors in the first sub-time interval after extraction of the liquid risk components of the illiquid risk factors;

(c) calculate a profit and loss (P&

L) distribution for the portfolio for the first sub-time interval [t₀, t₁] based on the residuals and the liquid risk factors for the first sub-time interval;

(d) repeat steps (a)-(c) for sub-time intervals indexed by k=2, . . . , m;

(e) compute a combined P&

L distribution for the portfolio by combining the P&

L distributions for each of the sub-time intervals indexed by k=1, . . . , m; and

(f) calculate the VaR for the portfolio for the n-day VaR calculation period based on the combined P&

L distribution.

25. The system of claim 24, wherein calculating the combined P&

L distribution comprises convolving the P&

L distributions calculated for k=1, . . . , m.

26. A non-transitory storage medium having stored thereon instructions that when executed by a processor circuit cause the processor circuit to calculate a liquidity-adjusted value-at-risk (LA-VaR) measure for a financial portfolio for a n-day VaR calculation period by:

where the n-day VaR calculation period has a number m of sub-time intervals indexed by k=1, . . . , m, each sub-time interval k having a start time t_k-1 and an end time t_k, and t₀=0,

(a) determining a core set of liquid risk factors available in a first sub-time interval [t₀, t₁], wherein the liquid risk factors available in the first sub-time interval are risk factors that have liquidity horizons less than or equal to the end time t₁ of the first sub-time interval and represent a space of liquid risks for the portfolio for the first sub-time interval, wherein risk factors with liquidity horizons greater than the end time t₁ of the first sub-time interval are illiquid risk factors in the first sub-time interval;

(b) extracting liquid risk components of the illiquid risk factors in the first sub-time interval [t₀, t₁] via linear regression of each illiquid risk factor on the core set of liquid risk factor in the first sub-time interval, wherein residuals of the linear regression represent the illiquid risk factors in the first sub-time interval after extraction of the liquid risk components of the illiquid risk factors;

(c) calculating a profit and loss (P&

L) distribution for the portfolio for the first sub-time interval [t₀, t₁] based on the residuals and the liquid risk factors for the first sub-time interval;

(d) repeating steps (a)-(c) for sub-time intervals indexed by k=2, . . . , m;

(e) computing a combined P&

L distribution for the portfolio by combining the P&

L distributions for each of the sub-time intervals indexed by k=1, . . . , m; and

(f) calculating the VaR for the portfolio for the n-day VaR calculation period based on the combined P&

L distribution.

27. The non-transitory storage medium of claim 26, wherein the computer readable medium has instructions stored thereon which when executed by the processor circuit causes the processor circuit to compute the combined P&

L distribution by convolving the P&

L distributions calculated for k=1, . . . , m.