Identifying industry sectors using statistical clusterization

• US 6,907,403 B1
• Filed: 07/13/2000
• Issued: 06/14/2005
• Est. Priority Date: 07/13/2000
• Status: Active Grant

• Current Assignee: C4cast.com Incorporated
• Original Assignee: C4cast.com Incorporated
• Inventors: Jennings, William P., Phillips, G. Michael, Findlay, M. Chapman III, Klein, Stephen A., Rice, Mark E.
• Primary Examiner(s): GARG, YOGESH C
• Application Number:
  US09/615,025
• Time in Patent Office: 1,797 Days
• Field of Search: 705/36, 345/440, 345/649, 706/25, 707/7
• 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 40/04 Trading; Exchange, e.g. stocks, commodities, derivatives or currency exchange
• CPC Class Codes: G06Q 40/06 Asset management; Financial planning or analysis

Abstract

Provided is a technique for classifying stocks into business sectors by calculating, for each of multiple exogenous variables, a measure of a tendency for a value of a stock to change as a result of a change in a data value for each such exogenous variable. The foregoing step is then repeated for each of several different stocks. Finally, the different stocks are grouped into different sectors based on similarities of such measures of tendency to change.

Claims

1. A method for classifying assets into business sectors, said method comprising:

(a) utilizing a computer to execute computer-executable process steps that include steps to;

(i) calculate, for each of plural exogenous variables, a measure of a tendency for a value of an asset to change as a result of a change in a data value for said each exogenous variable;

(ii) repeat step (i) for each of plural different assets; and

(iii) group said plural different assets into plural different sectors based on similarities of said measures of tendency to change across said plural exogenous variables;

(b) assessing at least one of;

(I) how statistics for at least one of the groups formed in step (iii) vary over time; and

(II) how one asset in a group formed in step (iii) compares to other assets in the same group; and

(c) purchasing or selling an asset based on the assessment made in step (b).

2. A method according to claim 1, wherein said measure of tendency to change comprises a measure of elasticity.

3. A method according to claim 1, wherein step (i) comprises:

(i-1) processing historical data for value of the asset and historical data values for said plural exogenous variables to obtain a price formula for estimating the value of the asset as a function of the exogenous variables; and

(i-2) taking a derivative of the price formula to obtain a formula expressing said tendency to change.

4. A method according to claim 3, wherein step (i-1) comprises performing a statistical regression technique.

5. A method according to claim 3, wherein said price formula is expressed as a truncated Taylor series expansion.

6. A method according to claim 1, wherein step (iii) comprises performing a statistical clustering technique whereby said plural different sectors are defined by clusters resulting from said statistical clustering technique.

7. A method according to claim 1, wherein step (iii) comprises performing a statistical regression technique.

8. A method according to claim 1, wherein said step (b) comprises a step of calculating are presentative characteristic of assets in a specific sector used in step (iii).

9. A method according to claim 8, wherein said step (b) further comprises a step of comparing a characteristic of a specific asset in said specific sector to the representative characteristic of assets in said specific sector.

10. A method according to claim 8, wherein said representative characteristic comprises an average return to assets in said specific sector.

11. A method according to claim 10, wherein said average return is calculated using a weighted average.

12. A method according to claim 1, further comprising a step of periodically repeating steps (i) through (iii).

13. A method according to claim 12, further comprising a step of tracking a position of a particular asset over time relative to its assigned sector.

14. A method according to claim 12, further comprising a step of tracking reclassification of a particular asset from a first sector to a second sector.

15. A method according to claim 1, wherein step (i) comprises determining a formula for calculating said measure of tendency to change, said formula being a function of said exogenous variables.

16. A method according to claim 15, wherein said computer-executable process steps further comprise steps to:

calculate plural samples of said measure of tendency to change using said formula for each of said plural different assets; and

use said samples in step (c) for grouping said plural different assets into said plural different sectors.

17. A method according to claim 16, wherein said measure of tendency to change is calculated in step (i) for each of the plural different assets using historical data values for said exogenous variables over a same period of time.

18. A method according to claim 17, wherein said samples are taken from a region of a multi-dimensional space defined by said exogenous variables in which the historical data values for said exogenous variables used in step (i) are clustered.

19. A method according to claim 1, wherein step (i) comprises a step to process historical data for value of the asset and historical data values for said plural exogenous variables to obtain a price formula for estimating the value of the asset as a function of the exogenous variables.

20. A method according to claim 19, wherein said price formula is obtained by performing neural network processing.

21. A method according to claim 20, wherein said measure of tendency to change is calculated by inputting different data values for the exogenous variables and observing how an output of said price formula changes as a result of small changes in the data values for the exogenous variables.

22. A method according to claim 19, wherein said price formula is obtained by using a genetic algorithm.

23. A method according to claim 1, wherein each of the plural different assets comprises a share of stock.

24. A method according to claim 1, wherein each of the plural different assets comprises a bond.

25. A method according to claim 1, wherein at least one of said plural exogenous variables is an interest rate.

26. A method according to claim 1, wherein at least one of said plural exogenous variables is at least one of a commodity price and a commodity index.

27. A method according to claim 1, wherein at least one of said plural exogenous variables is a number of housing starts.

28. A method according to claim 1, wherein at least one of said plural exogenous variables is an unemployment rate.

29. A method according to claim 1, wherein at least one of said plural exogenous variables is an inflation rate.

30. A method according to claim 1, wherein at least one of said plural exogenous variables is a measure of production.

31. A method according to claim 1, wherein at least one of said plural exogenous variables is a measure pertaining to at least one of inventory or sales.

32. An apparatus for classifying assets into business sectors, said apparatus comprising:

(a) means for calculating, for each of plural exogenous variables, a measure of a tendency for a value of an asset to change as a result of a change in a data value for said each exogenous variable;

(b) means for repeating the calculating performed by means (a) for each of plural different assets;

(c) means for grouping said plural different assets into plural different sectors based on similarities of said measures of tendency to change across said plural exogenous variables;

(d) means for assessing at least one of;

(I) how statistics for at least one of the groups formed by said means (c) vary over time; and

(II) how one asset in a group formed by said means (c) compares to other assets in the same group; and

(e) means for at least one of purchasing, selling or recommending the purchase or sale of an asset based on the assessment made by said means (d).

33. A computer-readable medium storing computer-executable process steps for classifying assets into business sectors, said process steps comprising steps to:

(a) calculate, for each of plural exogenous variables, a measure of a tendency for a value of an asset to change as a result of a change in a data value for said each exogenous variable;

(b) repeat step (a) for each of plural different assets;

(c) group said plural different assets into plural different sectors based on similarities of said measures of tendency to change across said plural exogenous variables;

(d) assessing at least one of;

(I) how statistics for at least one of the groups formed in step (c) vary over time; and

(II) how one asset in a group formed in step (c) compares to other assets in the same group; and

(e) purchasing, selling or recommending the purchase or sale of an asset based on the assessment made in step (d).

34. A computer-readable medium according to claim 33, wherein said measure of tendency to change comprises a measure of elasticity.

35. A computer-readable medium according to claim 33, wherein step (a) comprises steps to:

(a1) process historical data for value of the asset and historical data values for said plural exogenous variables to obtain a price formula for estimating the value of the asset as a function of the exogenous variables; and

(a2) take a derivative of the price formula to obtain a formula expressing said tendency to change.

36. A computer-readable medium according to claim 35, wherein step (a1) comprises a step to perform a statistical regression technique.

37. A computer-readable medium according to claim 35, wherein said price formula is expressed as a truncated Taylor series expansion.

38. A computer-readable medium according to claim 33, wherein step (c) comprises a step to perform a statistical clustering technique whereby said plural different sectors are defined by clusters resulting from said statistical clustering technique.

39. A computer-readable medium according to claim 33, wherein step (c) comprises a step to perform a statistical regression technique.

40. A computer-readable medium according to claim 33, further comprising a step to calculate a representative characteristic of assets in a specific sector used in step (c).

41. A computer-readable medium according to claim 40, further comprising a step to compare a characteristic of a specific asset in said specific sector to the representative characteristic of assets in said specific sector.

42. A computer-readable medium according to claim 40, wherein said representative characteristic comprises an average return to assets in said specific sector.

43. A computer-readable medium according to claim 42, wherein said average return is calculated using a weighted average.

44. A computer-readable medium according to claim 33, further comprising a step to periodically repeat steps (a) through (c).

45. A method according to claim 44, further comprising a step to track a position of a particular asset over time relative to its assigned sector.

46. A computer-readable medium according to claim 44, further comprising a step to track reclassification of a particular asset from a first sector to a second sector.

47. A computer-readable medium according to claim 33, wherein step (a) comprises a step to determine a formula for calculating said measure of tendency to change, said formula being a function of said exogenous variables.

48. A computer-readable medium according to claim 47, further comprising steps to:

calculate plural samples of said measure of tendency to change using said formula for each of said plural different assets; and

use said samples in step (c) for grouping said plural different assets into said plural different sectors.

49. A computer-readable medium according to claim 48, wherein said measure of tendency to change is calculated in step (a) for each of the plural different assets using historical data values for said exogenous variables over a same period of time.

50. A computer-readable medium according to claim 49, wherein said samples are taken from a region of a multi-dimensional space defined by said exogenous variables in which the historical data values for said exogenous variables used in step (a) are clustered.

51. A computer-readable medium according to claim 33, step (a) comprises a step to process historical data for value of the asset and historical data values for said plural exogenous variables to obtain a price formula for estimating the value of the asset as a function of the exogenous variables.

52. A computer-readable medium according to claim 51, wherein said price formula is obtained by performing neural network processing.

53. A computer-readable medium according to claim 52, wherein said measure of tendency to change is calculated by inputting different data values for the exogenous variables and observing how an output of said price formula changes as a result of small changes in the data values for the exogenous variables.

54. A computer-readable medium according to claim 51, wherein said price formula is obtained by using a genetic algorithm.

55. A computer-readable medium according to claim 33, wherein each of the plural different assets comprises a share of stock.

56. A computer-readable medium according to claim 33, wherein each of the plural different assets comprises a bond.

57. A computer-readable medium according to claim 33, wherein at least one of said plural exogenous variables is an interest rate.

58. A computer-readable medium according to claim 33, wherein at least one of said plural exogenous variables is at least one of a commodity price and a commodity index.

59. A computer-readable medium according to claim 33, wherein at least one of said plural exogenous variables is a number of housing starts.

60. A computer-readable medium according to claim 33, wherein at least one of said plural exogenous variables is an unemployment rate.

61. A computer-readable medium according to claim 33, wherein at least one of said plural exogenous variables is an inflation rate.

62. A computer-readable medium according to claim 33, wherein at least one of said plural exogenous variables is a measure of production.

63. A computer-readable medium according to claim 33, wherein at least one of said plural exogenous variables is a measure pertaining to at least one of inventory or sales.