Price Sensitivity Coefficient Calculation Using Shadow Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional cost-of-service based bidding systems in the parcel shipping industry lack the ability to factor market response and competitor characteristics into pricing decisions, leading to inaccurate calculations of price sensitivity coefficients.
Innovation Solution
The creation of shadow data to enhance historical bid data sets, allowing for more accurate logistic regression analysis and determination of price sensitivity coefficients through the addition of modified records based on bid outcomes, enabling a market response model to calculate the probability of winning a bid and determine optimal prices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If traditional cost-of-service based bidding systems are used, then pricing decisions can be made based on service costs, but the ability to factor market response and competitor characteristics into pricing decisions is lost
Solution Approach 1:
The system performs preliminary actions by collecting and storing market response data and competitor characteristics in advance through web crawlers and data gathering modules. This historical data is then used by the logistic regression model to calculate price sensitivity coefficients, allowing the pricing system to incorporate market intelligence without adding complexity to the core pricing algorithm.
Solution Approach 2:
The patent introduces a logistic regression model as an intermediary between raw market data and pricing decisions. This mathematical model acts as a mediator that processes market response information and competitor characteristics, transforming them into actionable price sensitivity coefficients that guide bid pricing without requiring complex integration of multiple data sources.
2Measurement precision
If logistic regression is used to calculate price sensitivity coefficient, then market response characteristics can be incorporated, but insufficient data leads to inaccurate calculations
Solution Approach 1:
The system performs preliminary data enrichment by using web crawlers to gather additional market data, competitor pricing information, and bid outcome data before running the logistic regression analysis. This pre-collection and expansion of historical bid data ensures sufficient data volume for accurate coefficient calculation.
Solution Approach 2:
The system implements feedback mechanisms where bid outcomes (won/lost) are fed back into the historical data set, and the logistic regression model is continuously retrained with this enriched data. This feedback loop progressively improves the accuracy of price sensitivity coefficients by learning from actual market responses.
Data Source
AI summary
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.