Predicting confidential data value insights at organization level using peer organization group

1. A system comprising:

a non-transitory computer-readable medium having instructions stored thereon, which, when executed by a processor, cause the system to:

determine that a first deidentified set of confidential data values for a first organization contains fewer values than a predetermined threshold for providing a statistical insight, the first deidentified set of confidential data submitted by users in a cohort having a first value for a first attribute and a second value for a second attribute;

in response to the determining:

retrieve a second deidentified set of confidential data values for organizations identified as peer organizations for the first organization, the second deidentified set of confidential data values submitted by users in a cohort having the first value for the first attribute and the second value for the second attribute;

retrieve a third deidentified set of confidential data values for the first organization, the third deidentified set of confidential data values submitted by users in a cohort having no limitation on values for the first attribute or the second attribute;

modeling the second deidentified set of confidential data values using a Bayesian model and smoothing results using the third deidentified set of confidential data values;

infer one or more confidential data values for the first organization using the modeled and smoothed second deidentified set of confidential data values;

calculate a statistical insight based on the inferred one or more confidential data values; and

cause display of the statistical insight in a graphical user interface rendered on a display.

2. The system of claim 1 , wherein the first attribute is title and the second attribute is region.

3. The system of claim 1 , wherein the confidential data values are compensation values for employment.

4. The system of claim 1 , wherein the inferring includes inferring a mean based on a mean of the third deidentified set and a residual whose variance is a combination of organization adjustment and random noise.

5. The system of claim 1 , wherein the instructions further cause the system to:

model the third deidentified set of confidential data values using a regression model.

6. The system of claim 5 , wherein the instructions further cause the system to combine output of the model of the second deidentified set of confidential data values and output of the model of the third deidentified set of confidential data values using a log-normal distribution assumption.

7. The system of claim 1 , wherein the instructions further cause the system to generate a graphical user interface containing the statistical insight as well a graphic depicting a histogram related to the statistical insight.

8. A computerized method, comprising:

determining that a first deidentified set of confidential data values for a first organization contains fewer values than a predetermined threshold for providing a statistical insight, the first deidentified set of confidential data submitted by users in a cohort having a first value for a first attribute and a second value for a second attribute;

in response to the determining:

retrieving a second deidentified set of confidential data values for organizations identified as peer organizations for the first organization, the second deidentified set of confidential data values submitted by users in a cohort having the first value for the first attribute and the second value for the second attribute;

retrieving a third deidentified set of confidential data values for the first organization, the third deidentified set of confidential data values submitted by users in a cohort having no limitation on values for the first attribute or the second attribute;

modeling the second deidentified set of confidential data values using a Bayesian model and smoothing results using the third deidentified set of confidential data values;

inferring one or more confidential data values for the first organization using the modeled and smoothed second deidentified set of confidential data values;

calculating a statistical insight based on the inferred one or more confidential data values; and

causing display of the statistical insight in a graphical user interface rendered on a display.

9. The method of claim 8 , wherein the first attribute is title and the second attribute is region.

10. The method of claim 8 , wherein the confidential data values are compensation values for employment.

11. The method of claim 8 , wherein the inferring includes inferring a mean based on a mean of the third deidentified set and a residual whose variance is a combination of organization adjustment and random noise.

12. The method of claim 8 , further comprising:

modeling the third deidentified set of confidential data values using a regression model.

13. The method of claim 12 , further comprising combining output of the model of the second deidentified set of confidential data values and output of the model of the third deidentified set of confidential data values using a log-normal distribution assumption.

14. The method of claim 8 , further comprising generating a graphical user interface containing the statistical insight as well a graphic depicting a histogram related to the statistical insight.

15. A non-transitory machine-readable storage medium comprising instructions, which when implemented by one or more machines, cause the one or more machines to perform operations comprising:

determining that a first deidentified set of confidential data values for a first organization contains fewer values than a predetermined threshold for providing a statistical insight, the first deidentified set of confidential data submitted by users in a cohort having a first value for a first attribute and a second value for a second attribute;

in response to the determining:

retrieving a second deidentified set of confidential data values for organizations identified as peer organizations for the first organization, the second deidentified set of confidential data values submitted by users in a cohort having the first value for the first attribute and the second value for the second attribute;

retrieving a third deidentified set of confidential data values for the first organization, the third deidentified set of confidential data values submitted by users in a cohort having no limitation on values for the first attribute or the second attribute;

modeling the second deidentified set of confidential data values using a Bayesian model and smoothing results using the third deidentified set of confidential data values;

inferring one or more confidential data values for the first organization using the modeled and smoothed second deidentified set of confidential data values;

calculating a statistical insight based on the inferred one or more confidential data values; and

causing display of the statistical insight in a graphical user interface rendered on a display.

16. The non-transitory machine-readable storage medium of claim 15 , wherein the first attribute is title and the second attribute is region.

17. The non-transitory machine-readable storage medium of claim 15 , wherein the confidential data values are compensation values for employment.

18. The non-transitory machine-readable storage medium of claim 17 , wherein the operations further comprise combining output of the model of the second deidentified set of confidential data values and output of the model of the third deidentified set of confidential data values using a log-normal distribution assumption.

19. The non-transitory machine-readable storage medium of claim 15 , wherein the inferring includes inferring a mean based on a mean of the third deidentified set and a residual whose variance is a combination of organization adjustment and random noise.

20. The non-transitory machine-readable storage medium of claim 15 , wherein the operations further comprise:

modeling the third deidentified set of confidential data values using a regression model.