Methods and apparatus to determine synthetic respondent level data using constrained Markov chains
Methods, apparatus, systems, and articles of manufacture are disclosed to generate synthetic respondent level data. Example apparatus disclosed herein include a synthetic panel generator to (i) generate a synthetic panel corresponding to a duration of time, the synthetic panel to be generated based on (a) a seed panel and (b) duration constraints based on return path data reported by a plurality of media devices, the synthetic panel representative of audiences of media presented by the plurality of media devices during the duration of time, and (ii) generate synthetic respondent level data based on viewing data of synthetic panelists in the synthetic panel. Disclosed example apparatus also include an output file generator to generate an output file based on the synthetic respondent level data.
1. An apparatus comprising:
a synthetic panel generator to:
generate a transition matrix corresponding to a first sub-duration of a duration of time and a second sub-duration of the duration of time, the second sub-duration different than the first sub-duration, the transition matrix to be based on a seed panel;
generate, based on the transition matrix, viewing data of synthetic panelists, the viewing data to correspond to the first and second sub-durations of time,
determine weights to apply to the viewing data of the synthetic panelists to satisfy duration constraints, the duration constraints based on return path data reported by a plurality of media devices;
remove ones of the synthetic panelists associated with one of the weights that do not satisfy a threshold to generate a synthetic panel corresponding to the duration of time, the synthetic panel representative of audiences of media presented by the plurality of media devices during the duration of time; and
generate synthetic respondent level data based on the viewing data of the synthetic panelists in the synthetic panel; and
an output file generator to generate an output file based on the synthetic respondent level data.
2. The apparatus of claim 1 , wherein the transition matrix corresponds to a probability that a panelist will change stations during a sub-duration.
3. The apparatus of claim 1 , wherein the synthetic panel generator is to reweight the synthetic panelists after the ones of the synthetic panelists have been removed.
4. The apparatus of claim 1 , wherein the duration constraints include reaches for the duration of time, the reaches determined based on the return path data.
5. The apparatus of claim 1 , wherein the synthetic respondent level data includes time records for viewing sessions by the synthetic panelists and corresponding demographic data.
6. The apparatus of claim 1 , wherein the output file includes the synthetic respondent level data.
7. A method comprising:
generating, by executing an instruction with a processor, a transition matrix corresponding to a first sub-duration of a duration of time and a second sub-duration of the duration of time, the second sub-duration different than the first sub-duration, the transition matrix to be based on a seed panel;
generating, by executing an instruction with the processor, viewing data of synthetic panelists based on the transition matrix, the viewing data to correspond to the first and second sub-durations of time,
determining, by executing an instruction with the processor, weights to apply to the viewing data of the synthetic panelists to satisfy duration constraints, the duration constraints based on return path data reported by a plurality of media devices;
removing, by executing an instruction with the processor, ones of the synthetic panelists associated with one of the weights that do not satisfy a threshold to generate a synthetic panel corresponding to the duration of time, the synthetic panel representative of audiences of media presented by the plurality of media devices during the duration of time;
generating, by executing an instruction with the processor, synthetic respondent level data based on the viewing data of the synthetic panelists in the synthetic panel; and
generating an output file based on the synthetic respondent level data.
8. The method of claim 7 , wherein the transition matrix corresponds to a probability that a panelist will change stations during a sub-duration.
9. The method of claim 7 , further including reweighting the synthetic panelists after the ones of the synthetic panelists have been removed.
10. The method of claim 7 , wherein the duration constraints include reaches for the duration of time, the reaches determined based on the return path data.
11. The method of claim 7 , wherein the synthetic respondent level data includes time records for viewing sessions by the synthetic panelists and corresponding demographic data.
12. The method of claim 7 , wherein the output file includes the synthetic respondent level data.
13. A non-transitory computer readable storage medium comprising instructions which, when executed, cause a machine to at least:
generate a transition matrix corresponding to a first sub-duration of a duration of time and a second sub-duration of the duration of time, the second sub-duration different than the first sub-duration, the transition matrix to be based on a seed panel;
generate, based on the transition matrix, viewing data of synthetic panelists, the viewing data to correspond to the first and second sub-durations of time,
determine weights to apply to the viewing data of the synthetic panelists to satisfy duration constraints, the duration constraints based on return path data reported by a plurality of media devices;
remove ones of the synthetic panelists associated with one of the weights that do not satisfy a threshold to generate a synthetic panel corresponding to the duration of time, the synthetic panel representative of audiences of media presented by the plurality of media devices during the duration of time;
generate synthetic respondent level data based on the viewing data of synthetic panelists in the synthetic panel; and
generate an output file based on the synthetic respondent level data.
14. The non-transitory computer readable storage medium of claim 13 , wherein the transition matrix corresponds to a probability that a panelist will change stations during a sub-duration.
15. The non-transitory computer readable storage medium of claim 13 , wherein the instructions, when executed, cause the machine to reweight the synthetic panelists after the ones of the synthetic panelists have been removed.
16. The non-transitory computer readable storage medium of claim 13 , wherein the duration constraints include reaches for the duration of time, the reaches determined based on the return path data.
17. The non-transitory computer readable storage medium of claim 13 , wherein the synthetic respondent level data includes time records for viewing sessions by the synthetic panelists and corresponding demographic data.
18. An apparatus comprising:
memory; and
processor circuitry to execute computer readable instructions to:
generate a transition matrix corresponding to a first sub-duration of a duration of time and a second sub-duration of the duration of time, the second sub-duration different than the first sub-duration, the transition matrix to be based on a seed panel;
generate, based on the transition matrix, viewing data of synthetic panelists, the viewing data to correspond to the first and second sub-durations of time,
determine weights to apply to the viewing data of the synthetic panelists to satisfy duration constraints, the duration constraints based on return path data reported by a plurality of media devices;
remove ones of the synthetic panelists associated with one of the weights that do not satisfy a threshold to generate a synthetic panel corresponding to the duration of time, the synthetic panel representative of audiences of media presented by the plurality of media devices during the duration of time;
generate synthetic respondent level data based on the viewing data of synthetic panelists in the synthetic panel; and
generate an output file based on the synthetic respondent level data.
19. The apparatus of claim 18 , wherein the transition matrix corresponds to a probability that a panelist will change stations during a sub-duration.
20. The apparatus of claim 18 , wherein the processor circuitry is to reweight the synthetic panelists after the ones of the synthetic panelists have been removed.
21. The apparatus of claim 18 , wherein the duration constraints include reaches for the duration of time, the reaches determined based on the return path data.
22. The apparatus of claim 18 , wherein the synthetic respondent level data includes time records for viewing sessions by the synthetic panelists and corresponding demographic data.