A Simple Approach to Forecasting Correlated Products

A SIMPLE APPROACH TO FORECASTING CORRELATED PRODUCTS By Larry Lapide (This is an ongoing column in The Journal, which is intended to give a brief view on a potential topic of interest to practitioners of business forecasting. Suggestions on topics that you would like to see covered should be sent via email to llapide@mit.edu Ed) During my tenure as a market analyst, I got briefed by a host of demand forecasting and planning software vendors. I found during that time that while they are all unique in their own way, to most people the vendors all look the same. For example, one area of functionality that I found differentiated them (since only a few software vendors addressed it) was forecasting demand for correlated products, where each product potentially impacts the demand of other products. Most vendors’ software only deal with forecasting a product’s demand in isolation of other products, assuming demand among products is statistically independent. This works well in most cases, but can backfire in cases where demand among products is significantly correlated. WHEN ARE PRODUCTS CORRELATED? There are a variety of situations when demand among some products depends on each ...