System and method for providing a common data and analytic framework for valuating financial instruments and portfolios

• US 7,792,714 B1
• Filed: 12/20/2002
• Issued: 09/07/2010
• Est. Priority Date: 12/28/2001
• Status: Active Grant

• Current Assignee: Fannie Mae
• Original Assignee: Fannie Mae
• Inventors: Shulman, Alexander L., Mills, Leonard O.
• Primary Examiner(s): Kazimi; Hani
• Assistant Examiner(s): Shrestha; Bijendra K
• Application Number:
  US10/324,097
• Time in Patent Office: 2,818 Days
• Field of Search: 705/35
• US Class Current: 705/35
• CPC Class Codes: G06Q 40/00 Finance; Insurance; Tax strategies; Processing of corporate or income taxes
• CPC Class Codes: G06Q 40/06 Asset management; Financial planning or analysis

Abstract

A distributed system that provides a common data and analytic model platform for enabling users of different applications to evaluate financial instruments. The system includes multiple user applications, and a computer network with multiple analytic servers that each includes simulation models for evaluating the financial instruments. The computer network also includes components which accept requests from the user applications, executes each request against an appropriate simulation model and returns results from the simulation model to the user application.

Claims

1. A distributed system that provides a common data and analytic model platform for enabling users of different applications to evaluate at least one financial instrument by accessing and executing at least one simulation model, the system comprising:

at least one user application in communication with a network;

the network comprising a plurality of components for accepting a plurality of requests for execution against different simulation models from the user application, each request, from the plurality of requests, including a financial instrument, a simulation model relating to the financial instrument, and a calculator;

at least one of the plurality of components for executing each request against the appropriate simulation model by executing an analytic server that includes the simulation model for evaluating the financial instrument and returning associated results from the simulation model to the user application, the simulation model being one of an interest rate model, a pre-payment model, loss severity model, finance rate model, a home price model, and a property valuation model,

wherein the plurality of components includes at least one enterprise server for receiving requests from the user application such that the enterprise server is a high availability server responsible for splitting up requests into independent tasks and at least one computational server cluster for accepting tasks from the enterprise server and providing specialized processing capabilities for executing simulation models such that the computational server cluster includes the analytic server,

wherein the analytic server including a parser that parses the request to determine the dependence between different calculations implemented on the request and parses the request into optimal calculations and a cache memory into which the parser stores previous calculations from the requests and dynamically reviews the results from the previous calculations to determine an optimal manner for executing each future calculation by substituting results from the calculations into the original request; and

at least one file server that includes storage media for storing a copy of the programming code and data used in the system.

2. The system of claim 1, wherein the computational server cluster comprises at least one specialized computing component in communications with the analytic server for evaluating the financial instruments.

3. The system of claim 1, wherein the programming code and data are stored in the file server in a static state.

4. The system of claim 1, wherein multiple historical versions on the programming code and data are stored in the file server.

5. The system of claim 1, wherein the enterprise server includes a local copy of the programming code and synchronization utilities ensures that the programming code on the file server and the enterprise server.

6. The system of claim 1, wherein the enterprise server receives an evaluation request from the user application, uses the evaluation request to determine the appropriate simulation model to be executed, and efficiently distributes the evaluation request and the associated processing requirements within the network.

7. The system of claim 1, wherein the user application is any application related to evaluating a mortgage portfolio.

8. The system of claim 1, wherein the plurality of components comprise:

a plurality of enterprise server for receiving requests from the user application, wherein each enterprise server is a high availability server responsible for splitting up requests into independent tasks and at least one load balancer is used to distribute requests between the plurality of enterprise servers;

at least one computational server cluster for accepting tasks from the enterprise server and providing specialized processing capabilities for executing simulation models, wherein the computational server cluster includes the analytic server; and

at least one file server that includes storage media for storing a copy of the programming code and data used in the system.

9. The system of claim 8, wherein if one of the plurality of enterprise servers fails, the load balancer redirects the requests to another one of the plurality of enterprise servers.

10. The system of claim 1, wherein each request is submitted to the network using text request.

11. The system of claim 1, wherein the financial instrument is a mortgage portfolio.

12. The system of claim 1, further comprising at least one optional field for submitting optional information to the network.

13. The system of claim 1, wherein the analytic server comprises:

at least one simulation library for storing a plurality of simulation models;

at least one configuration file which defines the parameters for how the at least one simulation algorithm implements the request against the simulation library; and

at least one application interface for enabling the plurality of simulation models to communicate with components in the system.

14. A method for enabling users of different applications to evaluate financial instruments by accessing and executing at least one simulation model in a distributed computer network, the method comprising the steps of

communicating a valuation request from at least one user application to a network;

accepting the valuation request from the user application by the network, the valuation request including a financial instrument, a simulation model, and a calculator;

dividing the valuation request into tasks, identifying the dependencies among the different tasks;

transferring the tasks to at least one component in communications with an appropriate simulation model;

executing each task against the appropriate simulation model by an analytic server, one task at a time;

caching each task result of executed tasks;

substituting task results from prior executed tasks in place of executing similar tasks;

returning results from the simulation model to the user application;

receiving requests from the user application by an enterprise server, wherein the enterprise server is a high availability server responsible for splitting up requests into independent tasks;

accepting tasks from the enterprise server by at least one computational server cluster to provide specialized processing capabilities for executing simulation models, wherein the computational server cluster includes the analytic server;

parsing the request by the analytic server to determine the dependence between different calculations implemented on the request and parsing the request into optimal calculations, storing previous calculations from the requests in a cache memory, and dynamically reviewing the results from the previous calculations to determine an optimal manner for executing each future calculation by substituting results from the calculations into the original request; and

storing a copy of the programming code and data used in the system by at least one file server.

15. The method of claim 14, wherein the valuation request comprises a plurality of financial instruments and the step of accepting further comprises the step of dividing the valuation request into sub requests.

16. A computer-readable medium whose contents cause a computer system to enable users of different applications to evaluate financial instruments by accessing and executing at least one simulation model in a distributed computer network, by performing the steps of:

communicating a valuation request from at least one user application to a network;

accepting the valuation request from the user application by the network, the valuation request comprising a plurality financial instrument, at least one simulation model, and at least one calculator;

determining the plurality of financial instruments comprising the valuation request and dividing the valuation request into sub requests;

dividing the valuation request and sub-requests into tasks;

transferring the tasks to at least one component in communications with an appropriate simulation model corresponding to at least one simulation model, the simulation model including one of an interest rate model, a pre-payment model, loss severity model, finance rate model, a home price model, and a property valuation model;

executing each task against the appropriate simulation model by an analytic server, wherein each analytic server processes one task at a time; and

returning results from the simulation model to the user application;

receiving requests from the user application by an enterprise server, wherein the enterprise server is a high availability server responsible for splitting up requests into independent tasks;

accepting tasks from the enterprise server by at least one computational server cluster to provide specialized processing capabilities for executing simulation models, wherein the computational server cluster includes the analytic server;

parsing the request by the analytic server to determine the dependence between different calculations implemented on the request and parsing the request into optimal calculations, storing previous calculations from the requests in a cache memory and dynamically reviewing the results from the previous calculations to determine an optimal manner for executing each future calculation by substituting results from the calculations into the original request; and

storing a copy of the programming code and data used in the system by at least one file server.

17. A method for enabling users of different applications to evaluate financial instruments by accessing and executing at least one simulation model in a distributed computer network, the method comprising the steps of:

communicating a valuation request from at least one user application to a network, the valuation request comprising a plurality financial instrument, at least one simulation model, and at least one calculator;

accepting the valuation request from the user application by the network;

dividing the valuation request into tasks;

transferring the tasks to at least one component in communications with an appropriate simulation model;

executing each task against the appropriate simulation model by analytic server, wherein each analytic server parses the request into at least one algorithm, accesses a configuration file for defining the parameter of how the at least one algorithm is to be executed, determines the dependence between different calculations to be implemented in the request, parses the request into optimal calculations by caching previous calculations in the request and dynamically reviewing the results from the previous calculations to determine an optimal manner for executing each upcoming calculation and processes one task at a time, and

returning results from the simulation model to the user application;

wherein the simulation model including one of an interest rate model, a pre-payment model, loss severity model, finance rate model, a home price model, and a property valuation model.

18. A distributed system that provides a common data and analytic model platform for enabling users of different applications to evaluate at least one financial instrument by accessing and executing at least one simulation model, the system comprising:

at least one user application in communication with a network; and

the network comprising;

at least one enterprise server for receiving requests from the user application and for executing the request against different simulation models, the request comprising a plurality financial instrument, at least one simulation model, and at least one calculator, and the enterprise server being a high availability server responsible for splitting up requests into independent tasks;

at least one computational server cluster for accepting tasks from the enterprise server and providing specialized processing capabilities for executing simulation models, wherein the computational server cluster includes the analytic server that executes each request against an appropriate simulation model that evaluates a financial instrument in the request and returns associated results from the simulation model to the user application, the simulation model including one of an interest rate model, a pre-payment model, loss severity model, finance rate model, a home price model, and a property valuation model;

the analytic server including a parser that parses the request to determine the dependence between different calculations implemented on the request and parses the request into optimal calculations and a cache memory into which the parser stores previous calculations from the requests and dynamically reviews the results from the previous calculations to determine an optimal manner for executing each future calculation by substituting results from the calculations into the original request; and

at least one file server that includes storage media for storing a copy of the programming code and data used in the system.