Methods, systems and apparatus to determine panel attrition

1. An audience measurement computing system for estimating panelist attrition, the audience measurement computing system comprising:

a processor;

at least one memory; memory having stored therein machine readable instructions that, when executed by the processor, cause the audience measurement computing system to perform operations comprising:

determining a beta distribution of a non-parametric survival curve estimate based on panel meter data associated with a cohort of panelists;

determining confidence intervals for a set of beta distribution parameters associated with the non-parametric survival curve estimate; and

outputting a panelist attrition estimate generated based on the confidence intervals, the panelist attrition estimate representing panelist retention over time based on an installation date associated with the panel meter data.

2. The audience measurement computing system of claim 1 , wherein the panel meter data includes at least one of a panel meter install date, a panel meter uninstall date, or activity log submission data.

3. The audience measurement computing system of claim 1 , wherein the operations further include applying a method of moments to estimate the set of beta distribution parameters when a variance of the non-parametric survival curve estimate is known.

4. The audience measurement computing system of claim 1 , wherein the non-parametric survival curve estimate matches a Kaplan-Meier point estimate.

5. The audience measurement computing system of claim 1 , wherein the operations further include determining the non-parametric survival curve estimate for censored and non-censored panel meter data.

6. The audience measurement computing system of claim 1 , wherein the panelist attrition estimate includes expected attrition rates for panelists based on a demographic attribute including at least one of an age, a gender, or a household size.

7. The audience measurement computing system of claim 1 , wherein the operations further include determining the non-parametric survival curve estimate based on a Haldane's prior distribution within the beta distribution.

8. A method for estimating panelist attrition, the method comprising:

determining a beta distribution of a non-parametric survival curve estimate based on panel meter data associated with a cohort of panelists;

determining confidence intervals for a set of beta distribution parameters associated with the non-parametric survival curve estimate; and

outputting a panelist attrition estimate generated based on the confidence, the panelist attrition estimate representing panelist retention over time based on an installation date associated with the panel meter data.

9. The method of claim 8 , wherein the panel meter data includes at least one of a panel meter install date, a panel meter uninstall date, or activity log submission data.

10. The method of claim 8 , further including applying a method of moments to estimate the beta distribution parameters when a variance of the non-parametric survival curve estimate is known.

11. The method of claim 8 , wherein the non-parametric survival curve estimate matches a Kaplan-Meier point estimate.

12. The method of claim 8 , further including determining the non-parametric survival curve estimate for censored and non-censored panel meter data.

13. The method of claim 8 , wherein the panelist attrition estimate includes expected attrition rates for panelists based on a demographic attribute including at least one of an age, a gender, or a household size.

14. The method of claim 8 , further including determining the non-parametric survival curve estimate based on a Haldane's prior distribution within the beta distribution.

15. A non-transitory computer readable storage medium having stored thereon instructions that, when executed by a processor of a computing system, cause the computing system to be configured to at least:

determine a beta distribution of a non-parametric survival curve estimate based on panel meter data associated with a cohort of panelists;

determine confidence intervals for a set of beta distribution parameters associated with the non-parametric survival curve estimate; and

output a panelist attrition estimate generated based on the confidence intervals, the panelist attrition estimate representing panelist retention over time based on an installation date associated with the panel meter data.

16. The non-transitory computer readable storage medium of claim 15 , wherein the instructions, when executed by the processor, cause the computing system to be configured to apply a method of moments to estimate the beta distribution parameters when a variance of the non-parametric survival curve estimate is known.

17. The non-transitory computer readable medium of claim 15 , wherein the non-parametric survival curve estimate matches a Kaplan-Meier point estimate.

18. The non-transitory computer readable medium of claim 15 , wherein the instructions, when executed by the processor, cause the computing system to be configured to determine the non-parametric survival curve estimate for censored and non-censored panel meter data.

19. The non-transitory computer readable medium of claim 15 , wherein the panelist attrition estimate includes expected attrition rates for panelists based on a demographic attribute including at least one of an age, a gender, or a household size.

20. The non-transitory computer readable medium of claim 15 , wherein the instructions, when executed by the processor, cause the computing system to be configured to determine the non-parametric survival curve estimate based on a Haldane's prior distribution within the beta distribution.