Systems and methods for improved calculation of coefficient for price sensitivity
AssigneeUnited Parcel Service of America, Inc.
InventorWarren Eastman Hearnes, II
Life of this patent
Estimating the term…
Abstract
Various embodiments of the present invention provide systems, methods, and computer program products for calculating a sufficiently accurate coefficient for price sensitivity to use in a target pricing system. In general, various embodiments of the invention involve providing an expanded data set by adding shadow data to a historical bid data set that allows a logistic regression approach to mathematically calculate the coefficient for price sensitivity with greater accuracy.
Claims11
A method for creating shadow data to enhance a historical bid data set, the historical bid data set comprising a plurality of records, each record of the plurality of records representing a bid for business, the method comprising the steps of:
- setting, via one or more processors, a variable N to a positive integer corresponding to a number of additional data records to be added for each record in the historical bid data set;
- setting, via the one or more processors, a variable Delta to a positive number corresponding to a percentage change in a price where each new data record is created;
- and
- for each record of the plurality of records:
- setting, via the one or more processors, a variable i equal to 0;
- setting, via the one or more processors, a variable P as an original price for a bid represented by the record;
- and
- looping, via the one or more processors, N number of times, wherein for each loop:
- in response to the bid being an accepted bid, modifying, via the one or more processors, the record to use a new variable P′ as a new price that is based upon a new price ratio of (PR−i*Delta);
- in response to the bid being a rejected bid, modifying, via the one or more processors, the record to use a new variable P′ as a new price that is based upon a new price ratio of (PR+i*Delta);
- saving, via the one or more processors, the modified record as a new record in the historical bid data set to form an enhanced historical bid data set;
- and
- adding, via the one or more processors, one to the variable i.
A system for creating shadow data to enhance a historical bid data set, the historical bid data set comprising a plurality of records, each record of the plurality of records representing a bid for business, the system comprising:
- memory;
- and
- at least one computer processor configured to:
- read a variable N from the memory, the variable N being a positive integer corresponding to a number of additional data records to be added for each record in the historical bid data set;
- read a variable Delta from the memory, the variable Delta being a positive number corresponding to a percentage change in a price where each new data record is created;
- and
- for each record of the plurality of records:
- set a variable i equal to 0;
- set a variable P as an original price for a bid represented by the record;
- and
- loop N number of times, wherein for each loop:
- in response to the bid being an accepted bid, modify the record to use a new variable P′ as a new price that is based upon a new price ratio of (PR−i*Delta);
- in response to the bid being a rejected bid, modify the record to use a new variable P′ as a new price that is based upon a new price ratio of (PR+i*Delta);
- store the modified record as a new record in the historical bid data set in the memory to form an enhanced historical bid data set;
- and
- add one to the variable i.
A non-transitory computer-readable medium containing executable code for creating shadow data to enhance a historical bid data set, the historical bid data set comprising a plurality of records, each record of the plurality of records representing a bid for business, that when executed by at least one computer processor causes the at least one computer processor to:
- read a variable N, the variable N being a positive integer corresponding to a number of additional data records to be added for each record in the historical bid data set;
- read a variable Delta, the variable Delta being a positive number corresponding to a percentage change in a price where each new data record is created;
- and
- for each record of the plurality of records:
- set a variable i equal to 0;
- set a variable P as an original price for a bid represented by the record;
- and
- loop N number of times, wherein for each loop:
- in response the bid being an accepted bid, modify the record to use a new variable P′ as a new price that is based upon a new price ratio of (PR−i*Delta);
- in response to the bid being a rejected bid, modify the record to use a new variable P′ as a new price that is based upon a new price ratio of (PR+i*Delta);
- store the modified record as a new record in the historical bid data set in the memory to form an enhanced historical bid data set;
- and
- add one to the variable i.
Dependencies are inferred from the claim text.