Methods and apparatus to determine weights for panelists in large scale problems

1. A method for determining a plurality of weights for a plurality of panelist representative of a population, the method comprising:

storing a constraint matrix and a population matrix into a memory, the constraint matrix representing first demographics associated with a plurality of panelists in a population, the population matrix representing second demographics of the population, the constraint matrix and population matrix corresponding to at least one of panelist data or population data obtained by a collector via a network communication;

determining, by executing an instruction with a processor, a first linear system based on a set of constraints including the constraint matrix and the population matrix;

determining, by executing an instruction with the processor, a first solution to the first linear system by determining an inverse of a first matrix of the first linear system, the determination of the inverse including breaking down the first matrix into blocks, the dimensions of the blocks being no larger than dimensions of the constraint matrix, the determination of the inverse including:

reducing the dimensions of a second element of a third matrix to a size less than or equal to the dimensions of the constraint matrix, the third matrix being an inversion of a second matrix, the second matrix being an inversion of the first matrix, the reduced second element being a third element;

storing the reduced third matrix in the memory;

replacing, in the stored third matrix, the second element with the third element in the second matrix; and

replacing a first element of the second matrix with the second matrix;

determining, by executing an instruction with the processor, a second linear system based on the first solution;

solving, by executing an instruction with the processor, the second linear system based on the set of constraints to determine an adjusted solution;

adjusting, by executing an instruction with the processor, an initial set of weights based on an inverse of a fourth matrix of the second linear system; and

outputting a resultant set of weights based on the inverse of the adjusted initial set of weights, the resultant set of weights satisfying the set of constraints.

2. The method of claim 1 , wherein the first linear system is a predictor linear system and the second linear system is a corrector linear system.

3. The method of claim 2 , wherein a second solution to the corrector linear system is calculated multiple times until the adjusted initial set of weight converges to the resultant set of weights, the multiple calculations based on prior solutions to the corrector linear system.

4. The method of claim 1 , further including applying the resultant set of weights to data associated with the panelists.

5. The method of claim 4 , wherein the applying of the resultant set of weights includes:

receiving a media exposure event from a panelist from the plurality of panelists with a weight from the set of weights; and

crediting demographics of the panelist for the media exposure event based on the weight.

6. The method of claim 1 , wherein the resultant set of weights is an optimal solution to an equality constraint, the equality constraint being an equation including the constraint matrix, the population matrix, and the set of weights.

7. The method of claim 6 , wherein:

the set of constraints further includes at least one of a non-negative constraint, a bound constraint, and the equality constraint; and

the optimal solution is a non-negative solution with minimum variance.

8. The method of claim 1 , wherein the inverse of the first matrix of the first linear system is generated by:

separating components of the first matrix into a first set of blocks;

inverting the first matrix using blockwise inversion to generate the second matrix;

separating a first element of the second matrix into a second set of blocks; and

inverting the second matrix using blockwise inversion to generate the third matrix, the second element of the third matrix having dimensions larger than the constraint matrix.

9. The method of claim 8 , wherein the dimensions of the second element are reduced using Sherman-Morrison formula.

10. The method of claim 9 , wherein the memory is conserved by eliminating a need to store the first linear system.

11. An apparatus to determine a plurality of weights for a plurality of panelist representative of a population, the apparatus comprising:

memory to store a constraint matrix and a population matrix, the constraint matrix representing first demographics associated with a plurality of panelists in a population, the population matrix representing second demographics of the population; and

a logic circuit to:

determine a first linear system based on a set of constraints including the constraint matrix and the population matrix;

determine a first solution to the first linear system by determining an inverse of a first matrix of the first linear system, the determination of the inverse including breaking down the first matrix into blocks, the dimensions of the blocks being no larger than dimensions of the constraint matrix, the determination of the inverse including:

reducing the dimensions of a second element of a third matrix to a size less than or equal to the dimensions of the constraint matrix, the third matrix being an inversion of a second matrix, the second matrix being an inversion of the first matrix, the reduced second element being a third element;

storing the reduced third matrix in the memory;

replacing, in the stored third matrix, the second element with the third element in the second matrix; and

replacing a first element of the second matrix with the second matrix;

determine a second linear system based on the first solution;

solve the second linear system based on the set of constraints to determine an adjusted solution;

adjust an initial set of weights based on an inverse of a fourth matrix of the second linear system; and

output a resultant set of weights based on the inverse of the adjusted initial set of weights, the resultant set of weights satisfying the set of constraints.

12. The apparatus of claim 11 , wherein a second solution to the second linear system is calculated multiple times until the adjusted initial set of weight converges to the resultant set of weights, the multiple calculations based on prior solutions to the second linear system.

13. The apparatus of claim 11 , wherein the logic circuit is to apply the resultant set of weights to data associated with the panelists.

14. The apparatus of claim 13 , wherein the logic circuit is to apply the resultant set of weights by:

receiving a media exposure event from a panelist from the plurality of panelists with a weight from the set of weights; and

crediting demographics of the panelist for the media exposure event based on the weight.

15. The apparatus of claim 11 , wherein the resultant set of weights is an optimal solution to an equality constraint, the equality constraint being an equation including the constraint matrix, the population matrix, and the set of weights.

16. The apparatus of claim 15 , wherein:

the set of constraints further includes at least one of a non-negative constraint, a bound constraint, and the equality constraint; and

the optimal solution is a non-negative solution minimum variance.

17. The apparatus of claim 11 , wherein the logic circuit is to determine the inverse of the matrix of the first linear system by:

separating components of the first matrix into a first set of blocks;

inverting the first matrix using blockwise inversion to generate the second matrix;

separating a first element of the second matrix into a second set of blocks; and

inverting the second matrix using blockwise inversion to generate the third matrix, the second element of the third matrix having dimensions larger than the constraint matrix.

18. The apparatus of claim 11 ,

wherein the logic circuit is further to:

obtain at least one of panelist data or population data via a network communication;

create the constraint matrix based on the panelist data; and

create the population matrix based on the population data.

19. A non-transitory computer readable medium comprising instructions that, when executed, cause a machine to at least:

store a constraint matrix and a population matrix into memory, the constraint matrix representing first demographics associated with a plurality of panelists in a population, the population matrix representing second demographics of the population;

determine a first linear system based on a set of constraints including the constraint matrix and the population matrix;

determine a first solution to the first linear system by determining an inverse of a first matrix of the first linear system, the determination of the inverse including breaking down the first matrix into blocks, the dimensions of the blocks being no larger than dimensions of the constraint matrix, the determination of the inverse including:

reducing the dimensions of a second element of a third matrix to a size less than or equal to the dimensions of the constraint matrix, the third matrix being an inversion of a second matrix, the second matrix being an inversion of the first matrix, the reduced second element being a third element;

storing the reduced third matrix in the memory;

replacing, in the stored third matrix, the second element with the third element in the second matrix; and

replacing a first element of the second matrix with the second matrix;

determine a second linear system based on the first solution;

solve the second linear system based on the set of constraints to determine an adjusted solution;

adjust an initial set of weights based on an inverse of a fourth matrix of a second linear system; and

output a resultant set of weights based on the inverse of the adjusted initial set of weights, the resultant set of weights satisfying the set of constraints.

20. The computer readable medium of claim 19 , wherein the first linear system is a predictor linear system and the second linear system is a corrector linear system.

21. The computer readable medium of claim 20 , wherein the instructions cause the machine to calculate a second solution to the corrector linear system multiple times until the adjusted initial set of weight converges to the resultant set of weights, the multiple calculations based on prior solutions to the corrector linear system.

22. The computer readable medium of claim 19 , wherein the resultant set of weights is an optimal solution to an equality constraint, the equality constraint being an equation including the constraint matrix, the population matrix, and the set of weights.

23. The computer readable medium of claim 22 , wherein:

the set of constraints further includes at least one of a non-negative constraint, a bound constraint, and the equality constraint; and

the optimal solution is a non-negative solution with minimum variance.

24. The computer readable medium of claim 19 , wherein the instructions cause the machine to determine the inverse of the first matrix of the first linear system by:

separating components of the first matrix into a first set of blocks;

inverting the first matrix using blockwise inversion to generate the second matrix;

separating a first element of the second matrix into a second set of blocks; and

inverting the second matrix using blockwise inversion to generate the third matrix, the second element of the third matrix having dimensions larger than the constraint matrix.