Methods and apparatus to model consumer choice sourcing

Abstract

Methods and apparatus are disclosed to model consumer choices. An example method includes identifying a set of products, receiving respondent choice data associated with the set of products, and adding the set of products to a base multinomial logit (MNL) model. The example method also includes generating, with a programmed processor, a number of copies of the MNL model to form an aggregate model based on a number of products in the set, each copy including an item utility parameter for each product in the set of products, and creating a matrix structure based on the number of products in the set, the matrix structure to be subtracted from each item utility parameter in the aggregate model. Further, the example method includes estimating each item utility parameter of the aggregate model and the matrix structure based on the number of copies of the MNL model and the respondent choice data, and calculating a choice probability based on each of the estimated utility parameters.

1 . A method to calculate a choice probability, comprising:
 identifying a set of products;   receiving respondent choice data associated with the set of products;   adding the set of products to a base multinomial logit (MNL) model;   generating, with a programmed processor, a number of copies of the MNL model to form an aggregate model based on a number of products in the set, each copy including an item utility parameter for each product in the set of products;   creating a matrix structure based on the number of products in the set, the matrix structure to be subtracted from each item utility parameter in the aggregate model;   estimating each item utility parameter of the aggregate model and the matrix structure based on the number of copies of the MNL model and the respondent choice data; and   calculating a choice probability based on each of the estimated utility parameters.   
     
     
         2 . (canceled) 
     
     
         3 . A method as described in  claim 1 , wherein the base MNL model size is based on the number of products in the set. 
     
     
         4 - 8 . (canceled) 
     
     
         9 . A method as described in  claim 1 , further comprising estimating the matrix structure with the aggregate model to facilitate parameter convergence of each item utility parameter and a plurality of parameters in the matrix structure. 
     
     
         10 . A method as described in  claim 1 , wherein the matrix structure comprises a straight matrix having a number of rows equal to the number of products and a number of columns equal to the number of products. 
     
     
         11 . A method as described in  claim 10 , wherein each of the number of rows represents one of the number of products and each of the number of columns represents one of the number of products. 
     
     
         12 . A method as described in  claim 11 , wherein an order of the products in the number of rows is the same as an order of the products in the number of columns. 
     
     
         13 . A method as described in  claim 12 , wherein each of the rows intersects each of the columns at a matrix cell to reflect a relationship between a product of the row with a product of the column. 
     
     
         14 . A method as described in  claim 13 , further comprising inserting a parameter at the matrix cell for each row and column intersection. 
     
     
         15 . A method as described in  claim 14 , wherein the parameter at the matrix cell for each row and column intersection is estimated with the respondent choice data to calculate converged parameter values therein. 
     
     
         16 . (canceled) 
     
     
         17 . (canceled) 
     
     
         18 . A method as described in  claim 1 , wherein the matrix structure comprises a geometric matrix having a number of rows equal to the number of products in the set, and a number of columns equal to a number of dimensions. 
     
     
         19 . (canceled) 
     
     
         20 . A method as described in  claim 18 , further comprising associating each cell within the geometric matrix with a spatial parameter indicative of a distance between a first and a second product from the number of products. 
     
     
         21 . A method as described in  claim 20 , wherein estimating further comprises converging a value of the spatial parameter based on the aggregate model and the respondent choice data. 
     
     
         22 . An apparatus to calculate a choice probability, comprising:
 a choice modeling engine to identify a set of products and receive respondent choice data associated with the set of products;   a multinomial logit (MNL) engine to add the set of products to a base MNL model;   an aggregate building engine to generate a number of copies of the MNL model to form an aggregate model based on a number of products in the set, each copy including an item utility parameter for each product in the set of products;   a sourcing modifier to create a matrix structure based on the number of products in the set, the matrix structure to be subtracted from each item utility parameter in the aggregate model;   an estimator to estimate each item utility parameter of the aggregate model and the matrix structure based on the number of copies of the MNL model and the respondent choice data; and   a simulation engine to calculate a choice probability based on each of the estimated utility parameters.   
     
     
         23 . (canceled) 
     
     
         24 . An apparatus as described in  claim 22 , wherein the MNL engine generates the base MNL model based on the number of products in the set. 
     
     
         25 . An apparatus as described in  claim 22 , further comprising the aggregate building engine inserting a price utility parameter for each product in the set of products. 
     
     
         26 - 29 . (canceled) 
     
     
         30 . An apparatus as described in  claim 22 , further comprising the estimator estimating the matrix structure with the aggregate model to facilitate parameter convergence of each item utility parameter and a plurality of parameters in the matrix structure. 
     
     
         31 . An apparatus as described in  claim 22 , further comprising a matrix engine to generate a straight matrix having a number of rows equal to the number of products and a number of columns equal to the number of products. 
     
     
         32 . An apparatus as described in  claim 31 , wherein each of the number of rows represents one of the number of products and each of the number of columns represents one of the number of products. 
     
     
         33 . An apparatus as described in  claim 32 , wherein the matrix engine places the number of products in the number of rows in the same order as the number of products in the number of columns. 
     
     
         34 . An apparatus as described in  claim 33 , wherein the matrix engine intersects each of the rows with each of the columns at a matrix cell to reflect a relationship between a product of the row with a product of the column. 
     
     
         35 . (canceled) 
     
     
         36 . (canceled) 
     
     
         37 . An apparatus as described in  claim 22 , further comprising a matrix spatial engine to generate a geometric matrix having a number of rows equal to the number of products in the set, and a number of columns equal to a number of dimensions. 
     
     
         38 . An apparatus as described in  claim 37 , wherein the matrix spatial engine associates each cell within the geometric matrix with a spatial parameter indicative of a distance between a first and a second product from the number of products. 
     
     
         39 . An apparatus as described in  claim 38 , wherein the estimator converges a value of the spatial parameter based on the aggregate model and the respondent choice data. 
     
     
         40 . A tangible machine accessible medium having instructions stored thereon that, when executed, cause a machine to, at least:
 identify a set of products;   receive respondent choice data associated with the set of products;   add the set of products to a base multinomial logit (MNL) model;   generate, with a programmed processor, a number of copies of the MNL model to form an aggregate model based on a number of products in the set, each copy including an item utility parameter for each product in the set of products;   create a matrix structure based on the number of products in the set, the matrix structure to be subtracted from each item utility parameter in the aggregate model;   estimate each item utility parameter of the aggregate model and the matrix structure based on the number of copies of the MNL model and the respondent choice data; and   calculate a choice probability based on each of the estimated utility parameters.   
     
     
         41 . (canceled) 
     
     
         42 . A tangible machine accessible medium as described in  claim 40  having instructions stored thereon that, when executed, cause a machine to generate the base MNL model having a size based on the number of products in the set. 
     
     
         43 . (canceled) 
     
     
         44 . (canceled) 
     
     
         45 . A tangible machine accessible medium as described in  claim 40  having instructions stored thereon that, when executed, cause a machine to include, in each item utility parameter, an attribute utility parameter. 
     
     
         46 . A tangible machine accessible medium as described in  claim 45  having instructions stored thereon that, when executed, cause a machine to include, in the attribute utility parameter, at least one of a product price utility, a product size utility, a product tradedress utility, or a product feature utility. 
     
     
         47 . (canceled) 
     
     
         48 . A tangible machine accessible medium as described in  claim 40  having instructions stored thereon that, when executed, cause a machine to estimate the matrix structure with the aggregate model to facilitate parameter convergence of each item utility parameter and a plurality of parameters in the matrix structure. 
     
     
         49 . A tangible machine accessible medium as described in  claim 40  having instructions stored thereon that, when executed, cause a machine to build a matrix structure as a straight matrix having a number of rows equal to the number of products and a number of columns equal to the number of products. 
     
     
         50 - 54 . (canceled) 
     
     
         55 . A tangible machine accessible medium as described in  claim 49  having instructions stored thereon that, when executed, cause a machine to convert the straight matrix to a symmetric matrix to constrain the estimation with the respondent choice data. 
     
     
         56 . (canceled) 
     
     
         57 . A tangible machine accessible medium as described in  claim 40  having instructions stored thereon that, when executed, cause a machine to generate the matrix structure as a geometric matrix having a number of rows equal to the number of products in the set, and a number of columns equal to a number of dimensions. 
     
     
         58 . A tangible machine accessible medium as described in  claim 57  having instructions stored thereon that, when executed, cause a machine to maintain the number of dimensions to be less than or equal to the number of products in the set. 
     
     
         59 . A tangible machine accessible medium as described in  claim 57  having instructions stored thereon that, when executed, cause a machine to associate each cell within the geometric matrix with a spatial parameter indicative of a distance between a first and a second product from the number of products. 
     
     
         60 . A tangible machine accessible medium as described in  claim 59  having instructions stored thereon that, when executed, cause a machine to converge a value of the spatial parameter based on the aggregate model and the respondent choice data.