Methods, systems and apparatus for calibrating data using relaxed benchmark constraints

Abstract

Methods, systems, and apparatus for calibrating data using relaxed benchmarks constraints are described. An example apparatus for generating a unique solution when calibrating data via a calibration model having relaxed benchmark constraints includes processor circuitry to execute computer-readable instructions. The computer-readable instructions are to execute the calibration model based on matrix data, a target loss function, a weight loss function, and a budget parameter. The computer-readable instructions are to determine calibrated weights resulting from execution of the calibration model. The computer-readable instructions are to incorporate a stability parameter into the calibration model in response to determining that the calibrated weights do not provide a unique solution for the executed calibration model. The stability parameter is to reduce an influence of the budget parameter on the calibration model to enable the generation of a unique solution.

What is claimed is: 
     
         1 . An apparatus for generating a unique solution when calibrating data via a calibration model having relaxed benchmark constraints, the apparatus comprising:
 at least one memory;   computer-readable instructions; and   processor circuitry to execute the computer-readable instructions to:
 execute the calibration model based on matrix data, a target loss function, a weight loss function, and a budget parameter, the matrix data based on criteria-based input data to be incorporated into the target loss function and unit-based input data to be incorporated into the weight loss function; 
 determine calibrated weights resulting from execution of the calibration model; and 
 incorporate a stability parameter into the calibration model in response to determining that the calibrated weights do not provide a unique solution for the executed calibration model, the stability parameter to reduce an influence of the budget parameter on the calibration model to enable the generation of a unique solution. 
   
     
     
         2 . The apparatus as defined in  claim 1 , wherein the stability parameter is to prevent a minimization procedure of the calibration model from reaching a flat region of the target loss function. 
     
     
         3 . The apparatus as defined in  claim 1 , wherein the matrix data is based on panelist data for a group of participants having one or more associated demographic representations, the criteria-based input data defined based on the one or more demographic representations. 
     
     
         4 . The apparatus as defined in  claim 1 , wherein the budget parameter is an upper constraint that a weight loss value determined via the weight loss function is allowed to obtain. 
     
     
         5 . The apparatus as defined in  claim 1 , wherein the processor circuitry is to:
 re-execute the calibration model based on the matrix data, the target loss function, the weight loss function, the budget parameter, and the stability parameter; and   re-determine the calibrated weights resulting from the re-execution of the calibration model, the re-determined calibrated weights to provide a unique solution for the re-executed calibration model.   
     
     
         6 . The apparatus as defined in  claim 1 , wherein the criteria-based input data includes (a) criteria variables and (b) at least one of targets for the criteria variables, upper and lower bounds for the criteria variables, scaling parameters for the criteria variables, or importance parameters for the criteria variables. 
     
     
         7 . The apparatus as defined in  claim 1 , wherein the unit-based input data includes (a) unit variables and (b) at least one of initial weights for the unit variables, upper and lower bounds for the unit variables, or size parameters for the unit variables. 
     
     
         8 . A tangible machine-readable storage medium comprising instructions that, when executed, cause a processor to at least:
 execute a calibration model based on matrix data, a target loss function, a weight loss function, and a budget parameter, the calibration model having relaxed benchmark constraints, the matrix data based on criteria-based input data to be incorporated into the target loss function and unit-based input data to be incorporated into the weight loss function;   determine calibrated weights resulting from the execution of the calibration model; and   incorporate a stability parameter into the calibration model in response to determining that the calibrated weights do not provide a unique solution for the executed calibration model, the stability parameter to reduce an influence of the budget parameter on the calibration model to enable the generation of a unique solution.   
     
     
         9 . The tangible machine-readable storage medium as defined in  claim 8 , wherein the stability parameter is to prevent a minimization procedure of the calibration model from reaching a flat region of the target loss function. 
     
     
         10 . The tangible machine-readable storage medium as defined in  claim 8 , wherein the matrix data is based on panelist data for a group of participants having one or more associated demographic representations, the criteria-based input data defined based on the one or more demographic representations. 
     
     
         11 . The tangible machine-readable storage medium as defined in  claim 8 , wherein the budget parameter is an upper constraint that a weight loss value determined via the weight loss function is allowed to obtain. 
     
     
         12 . The tangible machine-readable storage medium as defined in  claim 8 , wherein the instructions, when executed, are further to cause the processor to:
 re-execute the calibration model based on the matrix data, the target loss function, the weight loss function, the budget parameter, and the stability parameter; and   re-determine the calibrated weights resulting from the re-execution of the calibration model, the re-determined calibrated weights to provide a unique solution for the re-executed calibration model.   
     
     
         13 . The tangible machine-readable storage medium as defined in  claim 8 , wherein the criteria-based input data includes (a) criteria variables and (b) at least one of targets for the criteria variables, upper and lower bounds for the criteria variables, scaling parameters for the criteria variables, or importance parameters for the criteria variables. 
     
     
         14 . The tangible machine-readable storage medium as defined in  claim 8 , wherein the unit-based input data includes (a) unit variables and (b) at least one of initial weights for the unit variables, upper and lower bounds for the unit variables, or size parameters for the unit variables. 
     
     
         15 . A method for generating a unique solution when calibrating data via a calibration model having relaxed benchmark constraints, the method comprising:
 executing, by executing one or more computer readable instructions with a processor, the calibration model based on matrix data, a target loss function, a weight loss function, and a budget parameter, the matrix data based on criteria-based input data to be incorporated into the target loss function and unit-based input data to be incorporated into the weight loss function;   determining, by executing one or more computer readable instructions with the processor, calibrated weights resulting from the executing of the calibration model; and   incorporating, by executing one or more computer readable instructions with the processor, a stability parameter into the calibration model in response to determining that the calibrated weights do not provide a unique solution for the executed calibration model, the stability parameter to reduce an influence of the budget parameter on the calibration model to enable the generation of a unique solution.   
     
     
         16 . The method as defined in  claim 15 , wherein the stability parameter is to prevent a minimization procedure of the calibration model from reaching a flat region of the target loss function. 
     
     
         17 . The method as defined in  claim 15 , wherein the matrix data is based on panelist data for a group of participants having one or more associated demographic representations, the criteria-based input data defined based on the one or more demographic representations. 
     
     
         18 . The method as defined in  claim 15 , wherein the budget parameter is an upper constraint that a weight loss value determined via the weight loss function is allowed to obtain. 
     
     
         19 . The method as defined in  claim 15 , further including:
 re-executing the calibration model based on the matrix data, the target loss function, the weight loss function, the budget parameter, and the stability parameter; and   re-determining the calibrated weights resulting from the re-executing of the calibration model, the re-determined calibrated weights to provide a unique solution for the re-executed calibration model.   
     
     
         20 . The method as defined in  claim 15 , wherein the criteria-based input data includes (a) criteria variables and (b) at least one of targets for the criteria variables, upper and lower bounds for the criteria variables, scaling parameters for the criteria variables, or importance parameters for the criteria variables, and wherein the unit-based input data includes (c) unit variables and (d) at least one of initial weights for the unit variables, upper and lower bounds for the unit variables, or size parameters for the unit variables.