Methods and systems for forecasting product demand during promotional events using a causal methodology
An improved method for forecasting and modeling product demand for a product during promotional periods. The forecasting methodology employs a multivariable regression model to model the causal relationship between product demand and the attributes of past promotional activities. The model is utilized to calculate the promotional uplift from the coefficients of the regression equation. The methodology utilizes a mathematical formulation that transforms regression coefficients, a combination of additive and multiplicative coefficients, into a single promotional uplift coefficient that can be used directly in promotional demand forecasting calculations.
1. A method computer-implemented for forecasting product demand for a product during a future promotional period, the method comprising the steps of:
maintaining, in a data storage device, a database of historical product demand information;
calculating, by a computer in communication with said data storage device, a revised demand forecast for said product during said future promotional period from said historical product demand information;
identifying a plurality of causal factors influencing demand for said product during prior promotional periods;
analyzing, by said computer said historical product demand information for said product to determine a plurality of regression coefficients corresponding to said plurality of causal factors, said plurality of regression coefficients and corresponding causal factors being related through a multivariable regression equation: demand=a+b·promo k +c·decay+d·price; wherein promo k is a binary promotional flag for a media type k; decay is a binary flag indicating promotional decay; price, is a unit price for said product for a given week; and a, b, c, and d are said regression coefficients;
blending, by said computer, said plurality of regression coefficients to determine a single, multiplicative promotional uplift coefficient, said blending of said plurality of regression coefficients to determine a single, multiplicative promotional uplift coefficient comprising combining regression coefficients a, b, c, and d; and
combining, by said computer, said multiplicative promotional uplift coefficient with said revised demand forecast for said product to determine a promotional product demand for a product during a future promotional period.
2. A system for forecasting promotional demand for a product, comprising:
a data storage device containing a database of historical product demand information for a plurality of products; and
a computer in communication with said data storage device, said computer executing a product forecasting application for:
calculating a revised demand forecast for said product during said future promotional period from said historical product demand information;
identifying a plurality of causal factors influencing demand for said product during prior promotional periods;
analyzing said historical product demand information for said product to determine a plurality of regression coefficients corresponding to said plurality of causal factors, said plurality of regression coefficients and corresponding causal factors being related through a multivariable regression equation: demand=a+b·promo k +c·decay+d·price; wherein promo k is a binary promotional flag for a media type k; decay is a binary flag indicating promotional decay; price, is a unit price for said product for a given week; and a, b, c, and d are said regression coefficients;
blending said plurality of regression coefficients to determine a single, multiplicative promotional uplift coefficient, said blending of said plurality of regression coefficients to determine a single, multiplicative promotional uplift coefficient comprising combining regression coefficients a, b, c, and d; and
combining said multiplicative promotional uplift coefficient with said revised demand forecast for said product to determine a promotional product demand for a product during a future promotional period.