Using percentile data in business analysis of time series data

1. A real time data processing system for processing a stream of data events, comprising:

a computer device having:

a running profile processing system for updating a running profile each time a new data event value is inputted, wherein the new data event value an inputted financial transaction amount, and wherein the running profile includes percentile data that provides percentile ranges for associated financial transaction amounts;

an analysis system for analyzing the running profile, wherein the analysis system performs a composite analysis that utilizes: (a) results obtained from comparing a new data event value to the percentile data, and (b) stored results obtained from previous compare operations; and

outputting an analysis result relating to the inputted financial transaction amount.

2. The real time data processing system of claim 1 , wherein the percentile data is calculated using simulated percentiles.

3. The real time data processing system of claim 2 , wherein the simulated percentiles are calculated according to an algorithm:

if E n >X n-1 then X n =X n-1 +K , else X n =X n-1 −K,

wherein, E n is a current data event value, X n is a current simulated percentile,

X n-1 is a previously calculated simulated percentile, and K is an adjustment factor selected to generate a predetermined percentile.

4. The real time data processing system of claim 1 , wherein the percentile data comprises a histogram comprising a plurality of percentile ranges, each percentile defined by a boundary value, and each percentile range having a number of counts associated therewith.

5. The real time data processing system of claim 1 , wherein the analysis system outputs a warning if a predetermined condition for both a current compare operation and a previous compare operation are met.

6. The real time data processing system of claim 1 , wherein the stream of data events is selected from the group consisting of: financial data, business data, and network data.

7. A computer readable medium comprising:

a computer program product stored on the computer readable medium, which when executed, processes a stream of data events, the computer program product further comprising:

program code for updating a running profile each time a new data event value is inputted, wherein the new data event value includes an inputted financial transaction amount and wherein the running profile includes percentile data provides percentile ranges for associated financial transaction amounts;

program code for analyzing the running profile by performing a composite analysis that utilizes: (a) results obtained from comparing a new data event value to the percentile data, and (b) stored results obtained from previous compare operations; and

program code for outputting an analysis result relating to the inputted financial transaction amount.

8. The computer program product of claim 7 , wherein the percentile data is calculated using simulated percentiles.

9. The computer program product of claim 8 , wherein the simulated percentiles are calculated according to an algorithm:

if E n >X n-1 then X n =X n-1 +K , else X n =X n-1 −K,

wherein, E n is a current data event value, X n is a current simulated percentile,

X n-1 is a previously calculated simulated percentile, and K is an adjustment factor selected to generate a predetermined percentile.

10. The computer program product of claim 7 , wherein the percentile data comprises a histogram comprising a plurality of percentile ranges, each percentile defined by a boundary value, and each percentile range having a number of counts associated therewith.

11. The computer program product of claim 7 , wherein a warning is generated if a predetermined condition for both a current compare operation and a previous compare operation are met.

12. The computer program product of claim 7 , wherein the stream of data events is selected from the group consisting of: financial data, business data, and network data.

13. A method of processing a stream of data events, comprising;

utilizing a computer device to update a running profile each time a new data event value is inputted, wherein the new data event value includes a financial transaction and wherein the running profile includes percentile data that provides percentile ranges for associated financial transaction amounts; and

utilizing the computer device to analyze the running profile by performing a composite analysis that utilizes: (a) results obtained from comparing a new data event value to the percentile data, and (b) stored results obtained from previous compare operations.

utilizing the computer device to output an analysis result relating to the inputted financial transaction amount.

14. The method of claim 13 , wherein the percentile data is calculated using simulated percentiles.

15. The method of claim 14 , wherein the simulated percentiles are calculated according to an algorithm:

if E n >X n-1 then X n =X n-1 +K , else X n =X n-1 −K,

wherein, E n is a current data event value, X n is a current simulated percentile,

X n-1 is a previously calculated simulated percentile, and K is an adjustment factor selected to generate a predetermined percentile.

16. The method of claim 13 , wherein the percentile data comprises a histogram comprising a plurality of percentile ranges, each percentile defined by a boundary value, and each percentile range having a number of counts associated therewith.

17. The method of claim 13 , wherein a warning is generated if a predetermined condition for both a current compare operation and a previous compare operation are met.

18. The method of claim 13 , wherein the stream of data events is selected from the group consisting of: financial data, business data, and network data.