Merchant incremental electronic impact value prediction and ranking using multiple machine learning models

A computer-implemented method comprising:

• receiving a merchant market identifier associated with a total merchant data object set;
• receiving a target merchant identifier associated with a target merchant data object,
• wherein the total merchant data object set is associated with the target merchant data object and a remaining merchant data object set;
• determining a counterfactual input feature set associated with the remaining merchant data object set;
• identifying a counterfactual training market record set, the counterfactual training market record set associated with the counterfactual input feature set and a first electronic currency value set;
• training the counterfactual model based on the counterfactual training market record set;
• generating, based on the counterfactual input feature set and using a counterfactual model, a counterfactual electronic currency value associated with the remaining merchant data object set;
• identifying an actual electronic currency value associated with the total merchant data object set;
• determining a counterfactual incremental electronic currency impact associated with the target merchant data object based on the actual electronic currency value and the counterfactual electronic currency value;
• and
• training the counterfactual model based at least in part on the counterfactual electronic currency value or the counterfactual incremental electronic currency impact.

A computer program product comprising at least one non-transitory computer-readable storage medium having computer-executable program code instructions stored therein, the computer-executable program code instructions configured, in execution with a processor, to:

• receive a merchant market identifier associated with a total merchant data object set;
• receive a target merchant identifier associated with a target merchant data object,
• wherein the total merchant data object set is associated with the target merchant data object and a remaining merchant data object set;
• determine a counterfactual input feature set associated with the remaining merchant data object set;
• identifying a counterfactual training market record set, the counterfactual training market record set associated with the counterfactual input feature set and a first electronic currency value set;
• training the counterfactual model based on the counterfactual training market record set;
• generate, based on the counterfactual input feature set and using a counterfactual model, a counterfactual electronic currency value associated with the remaining merchant data object set;
• identify an actual electronic currency value associated with the total merchant data object set;
• determine a counterfactual incremental electronic currency impact associated with the target merchant data object based on the actual electronic currency value and the counterfactual electronic currency value;
• and
• train the counterfactual model based at least in part on the counterfactual electronic currency value or the counterfactual incremental electronic currency impact.