Systems and methods for measuring data quality over time

1. A computer-implemented method, performed by a computing platform connected to a database and a processor, for evaluating data quality, comprising:

storing, by the computing platform in the database, a plurality of records, the records sharing a common attribute;

reading, by the computing platform, first values for the common attribute corresponding to a first time period;

reading, by the computing platform, second values for the common attribute corresponding to a second time period;

accessing, by the computing platform and stored in the database, a business rule for evaluating the common attribute;

generating, by the computing platform, first consistency data for the common attribute, the first consistency data reflecting the extent to which the first values of the common attribute are consistent with the business rule at the first time period;

generating, by the computing platform, second consistency data for the common attribute, the second consistency data reflecting the extent to which the second values of the common attribute are consistent with the business rule at the second time period;

processing, by the computing platform and the processor, the first consistency data and the second consistency data to generate a quality change rate of the common attribute at a predetermined time period based on the first consistency data and the second consistency data;

determining, by the computing platform, third values of the common attribute at a future third time period based on the quality change rate;

generating, by the computing platform, an indication of whether the third values of the common attribute are below or above a threshold value;

determining, by the computing platform, a volatility measure of the common attribute by determining a number of changes to the common attribute from the first time period to the second time period;

comparing, by the computing platform, the volatility measure to a threshold volatility rate;

determining, by the computing platform, whether the volatility measure exceeds the threshold volatility rate;

determining, by the computing platform, a number of outcomes associated with the records; and

determining, by the computing platform and processor a relationship between the number of outcomes and the volatility measure based on whether the volatility measure exceeds the threshold volatility rate, wherein the records represent loans, and the outcomes are delinquencies or prepayments associated with the loans.

2. The computer-implemented method according to claim 1 , further comprising providing the quality change rate to the user as a graph or as a numerical value representing a percentage change in the data quality from the first time period to the second period.

3. The computer-implemented method according to claim 1 , further comprising:

providing a user with an interface for requesting the quality change rate, the interface including selectable options for identifying the business rule or values for the common attribute.

4. The computer-implemented method according to claim 3 , wherein the interface comprises a selectable option for aggregating the quality rate for a plurality of business rules.

5. The computer-implemented method according to claim 3 , wherein the interface comprises a selectable option for identifying a plurality of business rules.

6. The computer-implemented method according to claim 3 , wherein the interface comprises a selectable option for aggregating the quality rate for different values of the common attribute.

7. The computer-implemented method according to claim 3 , wherein the interface comprises a selectable option for identifying different values for a first attribute of the business records other than the common attribute, and individual quality rates for each identified value of the first attribute are provided to the user.

8. The computer-implemented method according to claim 7 , wherein the first attribute is an identifier of a lender.

9. The computer-implemented method according to claim 7 , wherein the first attribute represents a time period.

10. The computer-implemented method according to claim 1 , further comprising: determining an expected quality change rate for the common attribute.

11. The computer-implemented method according to claim 10 , wherein the expected quality change rate is determined using a mathematical technique comprising one or more of time series analysis, logistic regression, exponential regression, or probit regression.

12. The computer-implemented method according to claim 1 , further comprising:

determining a first error rate based on the extent to which the first values of the common attribute are consistent with the business rule at the first time;

comparing the first error rate to a threshold error rate; and

determining whether the first error rate exceeds the threshold error rate.

13. The computer-implemented method according to claim 12 , further comprising automatically determining a value for the threshold error rate based on previous error rates for the common attribute.

14. The computer-implemented method according to claim 1 wherein an aggregate number of changes are determined by accessing a change log reflecting changes to the common attribute from the first time period to the second time period.

15. The computer-implemented method according to claim 14 , wherein determining the aggregate number of changes comprises determining whether the common attribute changed by more than a predetermined amount.

16. The computer-implemented method according to claim 15 , wherein the predetermined amount is a fixed percentage of a value of the common attribute.

17. The computer-implemented method according to claim 1 , further comprising providing the volatility measure to the user as a numerical value or as a graph representing a percentage of the records for which the common attribute changed values from the first time period to the second period.

18. The computer-implemented method according to claim 1 , further comprising:

determining an expected volatility rate for the common attribute.

19. The computer-implemented method according to claim 1 , wherein the outcomes further represent values or data quality of other attributes of the records.

20. A system comprising:

a computer platform connected to a database and a processor; and

a computer-readable medium comprising instructions executable by the processor to:

read, by the computing platform, first values for a common attribute of a plurality of records, the first values corresponding to a first time period;

read, by the computing platform, second values for the common attribute corresponding to a second time period;

access, by the computing platform and stored in the database, a business rule for evaluating the common attribute;

generate, by the computing platform, first consistency data for the common attribute, the first consistency data reflecting the extent to which the first values of the common attribute are consistent with the business rule at the first time period;

generate, by the computing platform, second consistency data for the common attribute, the second consistency data reflecting the extent to which the second values of the common attribute are consistent with the business rule at the second time period;

process, by the computing platform and the processor, the first consistency data and the second consistency data to generate a quality change rate of the common attribute at a predetermined time period the first consistency data and the second consistency data;

determine, by the computing platform, third values of the common attribute at a future third time period based on the quality change rate;

generate, by the computing platform, an indication of whether the third values of the common attribute are below or above a threshold value;

determine, by the computing platform, a volatility measure of the common attribute by determining a number of changes to the common attribute from the first time period to the second time period;

compare, by the computing platform, the volatility measure to a threshold volatility rate;

determine, by the computing platform, whether the volatility measure exceeds the threshold volatility rate;

determine, by the computing platform, a number of outcomes associated with the records; and

determine, by the computing platform and processor, a relationship between the number of outcomes and the volatility measure based on whether the volatility measure exceeds the threshold volatility rate, wherein the records represent loans, and the outcomes are delinquencies or prepayments associated with the loans.

21. A non-transitory computer-readable medium storing processor-readable instructions, which when executed by a computing platform connected to a database and a processor, perform a method comprising:

reading, by the computing platform, first values for a common attribute of a plurality of records, the first values corresponding to a first time period;

reading, by the computing platform, second values for the common attribute corresponding to a second time period;

accessing, by the computing platform and stored in the database, a business rule for evaluating the common attribute;

generating, by the computing platform, first consistency data for the common attribute, the first consistency data reflecting the extent to which the first values of the common attribute are consistent with the business rule at the first time period;

generating, by the computing platform, second consistency data for the common attribute, the second consistency data reflecting the extent to which the second values of the common attribute are consistent with the business rule at the second time period; and

processing, by the computing platform and the processor, the first consistency data and the second consistency data to generate a quality change rate of the common attribute at a predetermined time period based on the first consistency data and the second consistency data;

determining, by the computing platform, third values of the common attribute at a future third time period based on the quality change rate;

generating, by the computing platform, an indication of whether the third values of the common attribute are below or above a threshold value;

determining, by the computing platform, a volatility measure of the common attribute by determining a number of changes to the common attribute from the first time period to the second time period;

comparing, by the computing platform, the volatility measure to a threshold volatility rate;

determining, by the computing platform, whether the volatility measure exceeds the threshold volatility rate;

determining, by the computing platform, a number of outcomes associated with the records; and

determining, by the platform and processor, a relationship between the number of outcomes and the volatility measure based on whether the volatility measure exceeds the threshold volatility rate wherein the records represent loans and the outcomes are delinquencies or prepayments associated with the loans.