Computer system for generating projected data and an application supporting a financial transaction

• US 6,456,982 B1
• Filed: 07/01/1993
• Issued: 09/24/2002
• Est. Priority Date: 07/01/1993
• Status: Expired due to Fees

• Current Assignee: Dragana N. Pilipovic
• Original Assignee: Dragana N. Pilipovic
• Inventors: Pilipovic, Dragana N.
• Primary Examiner(s): Sough, Hyung-Sub
• Assistant Examiner(s): Hewitt, II, Calvin L.
• Application Number:
  US08/086,002
• Time in Patent Office: 3,372 Days
• Field of Search: 364/401, 364/402, 364/408, 705/1, 705/7, 705/35, 705/36, 705/37, 705/10, 705/42
• US Class Current: 705/36.00R
• CPC Class Codes: G06Q 40/00 Finance; Insurance; Tax strategies; Processing of corporate or income taxes
• CPC Class Codes: G06Q 40/02 Banking, e.g. interest calculation or account maintenance credit or loans G06Q40/03
• 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

In a computer system, a method for making the system and a method for using the system, the invention includes providing a data processing system to generate projected data for variables. The data processing system includes a digital computer performing steps of processing the input data to calculate projected data respectively for a plurality of variables, and generating output including the projected data; wherein the processing was first tested for accuracy by preprocessing input test data to calculate projected test data for the variables and by preprocessing the projected test data to derive a portion of the input test data from the projected test data.

Claims

1. A method for using an apparatus to generate output by processing input market data in accordance with factors defined by testing for accuracy, the method including the steps of:

controlling a data processing system apparatus comprising a digital electrical computer with the processor operably connected to memory, means for receiving input data, and means for outputting processed market data, the processor controlled with logic means so that the system performs a step of processing input data entered at the means for inputting data to calculate projected market data respectively for each of a plurality of variables, and a step of generating output including the projected data at the means for outputting processed data;

wherein

the step of controlling includes directing the logic means with factors defined by testing for accuracy by preprocessing at least a portion of input test market data entered at the means for inputting data to calculate projected test data for each of the variables and preprocessing the projected test data to derive processed data equal to a portion of the input test data; and

the step of processing includes computing $V<br><br>\left[1,n\right]=\sqrt{n*{\left(M<br><br>\left[n\right]\right)}^2-\underset{k=1}{\overset{n-1}{\sum <br><br>}}<br><br>{\left(V<br><br>\left[n-k+1,k\right]\right)}^2},$

wherein V represents a discrete volatility matrix and is a portion of the projected data, M is a vector representing a portion of the input data, n represents an index number and k represents a summation coefficient.

2. The method of claim 1, wherein the input test data includes a liquid financial product market price;

and wherein the step of preprocessing the projected test data includes preprocessing the projected test data to derive the liquid financial product market price.

3. The method of claim 1, wherein the input data includes a liquid financial product market price;

and wherein the step of controlling further includes the step of;

processing the projected data, including projected data representing a projected liquid financial product price, and deriving an illiquid financial product expected price from the projected data.

4. The method of claim 1, wherein the input data includes price data;

and wherein the step of controlling the digital computer with logic means includes;

processing the input data to compute discrete forward price volatilities across forward prices and future calendar periods; and

wherein;

a portion of the projected data represents discrete forward price volatilities.

5. The method of claim 1, wherein the step of processing includes:

6. The method of claim 5, wherein the step of processing includes:

7. The method of claim 1, wherein some of the variables represent driver factors used in the processing.

8. The method of claim 7, wherein the step or preprocessing the projected test data include using the driver factors to derive the portion of the input test data from the projected test data.

9. The method of claim 7, wherein the step of controlling the digital computer with logic means further comprises forming a simulation of future behavior for each said driver factor to produce a corresponding driver factor distribution of values.

10. The method of claim 9 wherein one of the variables is a dependent variable;

and wherein the step of controlling the digital computer with logic means further comprises processing the values of at leas one of the driver factor distributions to form a simulation of future behavior of the dependent variable represented by a distribution of values for the dependent variable.

11. The method of claim 10, wherein the step of processing the values for the dependent variable includes processing such that an increased number of the values in the distribution of values for the dependent variable produces increased convergence on the portion of input data.

12. The method of claim 10, wherein the step of processing produces the output data including confidence intervals at the 66th percentile, the 95th percentile, and the 99th percentile.

13. The method of claim 10, wherein the dependent variable represents a liquid financial product price.

14. The method of claim 13, wherein the input data includes a liquid financial product market price;

and wherein the step of processing the values for the dependent variable includes processing such that an increased number of the values in the distribution of values for the dependent variable produces increased convergence on the liquid financial product market price.

15. The method of claim 10, wherein the dependent variable represents a illiquid financial product price.

16. The method of claim 15, wherein the input data includes a liquid financial product market price;

and wherein the step of processing the values for the dependent variable includes processing such that an increased number of the values in the distribution of values for the dependent variable produces increased convergence on the illiquid financial product expected price.

17. The method of claim 7, wherein the input data includes market forward prices;

and wherein some of the driver factors represent forward prices.

18. The method of claim 17, wherein one of the variables represents a liquid financial product price.

19. The method of claim 18, wherein the step of processing includes applying the driver factors to form a distribution of the values for the liquid financial product price.

20. The method of claim 17, wherein one of the variables represents an illiquid financial product price.

21. The method of claim 20, wherein the step of processing includes applying the driver factors to form a distribution of the values for the illiquid financial product price.

22. The method of claim 1, wherein the input data includes price data;

and wherein;

the step of controlling the digital computer with logic means includes processing the input data into discrete future forward prices across future calendar periods; and

wherein;

a portion of the projected data represents the future forward prices.

23. The method of claim 22, wherein the input data further includes a function representing cash flow of a financial product across the future calendar periods;

and wherein the step of controlling the digital computer with logic means includes the steps of;

generating forward price distributions from the projected data;

generating a cash flow distribution from the forward price distributions; and

calculating a price for the financial product from the generated cash flow distribution.

24. The method of claim 23, wherein the input function represents cash flow for an illiquid financial product.

25. The method of claim 23, wherein the step of controlling the digital computer with logic means further includes computing a delta value for the financial product;

and wherein the output includes the computed delta value.

26. The method of claim 23, wherein the input function represents cash flow for a portfolio of financial products.

27. The method of claim 26, wherein the step of controlling the digital computer with logic means further includes calculating a mean replacement cost of the portfolio;

and wherein the output includes the mean replacement cost.

28. The method of claim 23, wherein the financial product is an index amortizing swap;

and wherein the step of controlling the digital computer with logic means further includes the step of computing a break even price for the index amortizing swap; and

wherein the output includes the break even price.

29. The method of claim 28, further comprising the step of adding a spread to the break even price to produce a price quote for the index amortizing swap.

30. The method of claim 23, wherein the financial product is represented by swaption volatility;

and wherein the step of controlling the digital computer with logic means further includes the step of computing a break even price for the financial product represented by the swaption volatility; and

wherein the output includes the break even price.

31. The method of claim 30, further comprising the step of adding a spread to the break even price to produce a price quote for the financial product.

32. The method of claim 30, wherein the step of controlling the digital computer with logic means further includes the step of computing a delta value for the financial product represented by the swaption volatility;

and wherein the output includes the delta valve.

33. The method of claim 32, wherein the step of controlling the digital computer with logic means further includes the step of generating a remaining distribution after changing the portfolio by adding a financial product to the portfolio or by removing a financial product from the portfolio.

34. The method of any one of claims 1-33 further comprising the step of:

making a buy/sell/keep decision in response to the output, such that if the projected data includes a datum greater than a sale price for a financial product then making a buy decision, and if the projected data includes a datum less than the sale price then making a sell decision, and otherwise making a keep decision.

35. A method for using an apparatus to generate output by processing market data in accordance with factors defined by testing for accuracy, the method including the step of:

controlling a data processing system apparatus comprising a digital electrical computer with the processor operably connected to memory, means for receiving market data, and means for outputting processed market data, the processor controlled with logic means so that the system performs a step of processing input market data entered at the means for inputting market data to calculate projected market data respectively for each of a plurality of variables, and a step of generating output including the projected market data at the means for outputting processed market data;

wherein the step of controlling includes directing the logic means with factors defined by testing for accuracy by preprocessing at least a portion of input test market data entered at the means for inputting market data to calculate projected test market data for each of the variables and preprocessing the projected test market data to derive processed market data equal to a portion of the input test market data;

wherein some of the variables represent driver factors used in the processing, and the step of controlling further comprises forming a simulation of future behavior for each said driver factor to produce a corresponding driver factor distribution of values;

the variables include a plurality of dependent variables; and

wherein the step of controlling the digital computer with logic means further comprises;

processing the values of at least one of the driver factor distributions to form a simulation of future behavior of the dependent variables represented by distributions of values for each said dependent variable; and

wherein the step of processing the values for each said dependent variable includes;

a calculating an average value for each said distribution such that an increased number of the values in the distribution of values for the dependent variable produces increased convergence on the portion of the input market data;

36. The method of claim 35, wherein the variables include volatilities;

and

wherein the step of controlling the digital computer with logic means further comprises;

processing the projected data to form distributions of the volatilities, each distribution of the volatilities having an average, the distributions of the volatilities being generated so that the averages of the distributions of volatilities converge to the volatilities within the input data with increased numbers or the volatilities generated; and

wherein the output includes the distributions of volatilities.

37. The method of claim 35, wherein the input data includes market forward prices, market option volatilities, historical forward price volatilities, and historical correlations between the forward prices.

38. A data processing system apparatus for generating output by processing defined by testing, the output including projected market data for variables, the data processing system including:

a digital electrical computer having a processor operably connected to memory, to means for receiving input market data, and to means for outputting processed market data; and

means for programming the digital electrical computer to process input-data, entered at the means for inputting data, to calculate projected data respectively for each of a plurality of variables, and for generating output including the projected data at the means for outputting processed data;

wherein

the means for programming is made by a process including adjusting the means for programming and then testing for accuracy by preprocessing input test market data, entered at the means for inputting data, to calculate projected test data for each of the variables and followed by preprocessing the projected test data, the adjusting being carried out until the means for programming is adjusted sufficiently to derive processed data that converges to a portion of the input test data; and

said processing includes computing $V<br><br>\left[1,n\right]=\sqrt{n*{\left(M<br><br>\left[n\right]\right)}^2-\underset{k=1}{\overset{n-1}{\sum <br><br>}}<br><br>{\left(V<br><br>\left[n-k+1,k\right]\right)}^2},$

wherein V represents a discrete volatility matrix and is a portion of the projected data, M is a vector representing a portion of the input data, n represents an index number and k represents a summation coefficient.

39. An article of manufacture for making a data processing system generate projected market data for variables, the article of manufacturing including:

a disk storing logic means readable by a data processing system comprising a digital electrical computer having a processor operably connected to a disk drive, and operable for controlling the digital electrical computer to perform a step of processing input data, entered at the means for inputting data, to calculate projected market data respectively for each of a plurality of variables, and the step of generating output including the projected data at the means for outputting processed market data;

wherein

the logic means is made by a process including testing the processing for accuracy by preprocessing input test market data, entered at the means for inputting data, to calculate projected test data for each of the variables and preprocessing the projected test data until the processing defines factors for the logic means that derive a portion of the input test data from the projected test data; and

said processing includes computing $V<br><br>\left[1,n\right]=\sqrt{n*{\left(M<br><br>\left[n\right]\right)}^2-\underset{k=1}{\overset{n-1}{\sum <br><br>}}<br><br>{\left(V<br><br>\left[n-k+1,k\right]\right)}^2},$

wherein V represents a discrete volatility matrix and is a portion of the projected data, M is a vector representing a portion of the input data, n represents an index number and k represents a summation coefficient.

40. A computer-implemented method for using a data processing system apparatus comprising a digital electrical computer with the processor operably connected to memory, means for receiving input market data, and means for outputting processed market data, the method including the steps of:

processing, by logic means direction of processor operations, input historical market data entered at the means for inputting data into processed market data, by substeps including creating a three-dimensional framework using a statistical iterative search process to break down the input market data into time-sensitive building blocks;

using the building blocks in calibrating implied volatility parameter values functionally related to mathematically expressed econometric relationships, articulating end conditions to bound the statistical iterative search process, and generating data for each of a plurality of variables until the processed data converge on the input historical data;

projecting future data with the three-dimensional framework subsequent to the calibrating; and

generating output including the future market data at the means for outputting processed data;

wherein the step of processing includes computing $V<br><br>\left[1,n\right]=\sqrt{n*{\left(M<br><br>\left[n\right]\right)}^2-\underset{k=1}{\overset{n-1}{\sum <br><br>}}<br><br>{\left(V<br><br>\left[n-k+1,k\right]\right)}^2},$

wherein V represents a discrete volatility matrix and is a portion of the projected data, M is a vector representing a portion of the input data, n represents an index number and k represents a summation coefficient.

41. A computer-implemented method for using a data processing system apparatus comprising a digital electrical computer with the processor operably connected to memory, means for receiving input market data, and means for outputting processed market data, the method including the steps of:

processing, by, logic means direction of processor operations, input historical data entered at the means for inputting data into processed market data, by substeps including creating a three-dimensional framework utilizing the input historical data to articulate parameter values by breaking down the input historical data into building blocks with values, and by comparing the building block values so that the parameter values are calibrated for each cell of the three-dimensional framework to project market data by mathematically-expressed econometric relationships;

projecting future data with the three-dimensional framework subsequent to the calibrating; and

generating output including the future market data at the means for outputting processed data;

wherein the step of processing includes computing $V<br><br>\left[1,n\right]=\sqrt{n*{\left(M<br><br>\left[n\right]\right)}^2-\underset{k=1}{\overset{n-1}{\sum <br><br>}}<br><br>{\left(V<br><br>\left[n-k+1,k\right]\right)}^2},$

wherein V represents a discrete volatility matrix and is a portion of the projected data, M is a vector representing a portion of the input data, n represents an index number and k represents a summation coefficient.