October 16, 2019
Research Finds A Cheaper Solution to Predict Consumer Demand
Consumers are often influenced by several factors when making a purchase. When buying a smartphone, a consumer might consider price, screen size, operating system, ease of use, available data plans, or any number of other variables. This makes sense in terms of how a consumer thinks, but it complicates how the market predicts future demands. Apple and Samsung don’t release a new smartphone each time a customer discovers an alternative data plan. Predicting which variables will influence consumer demand is difficult and often costly, however, researchers found that it is possible to estimate how consumers discount the future versus the present.
The study, Linear Estimation of Aggregate Dynamic Discrete Demand for Durable Goods: Overcoming the Curse of Dimensionality, conducted by researchers at the Tepper School of Business at Carnegie Mellon University, Yale University and University of Leicester, was recently published by INFORMS Marketing Science.
“Previously, the focus was on data based on individual choices,” said Tim Derdenger, Associate Professor of Marketing and Strategy at the Tepper School, who coauthored the study. “Our focus is on technology products—durable goods—where prices fluctuate over time, and more likely actually decline as the product gets older. We wanted to find a way to cheaply estimate demand models for durable goods, allowing for consumers to form beliefs about what the future might hold for that product.”
The researchers developed a new approach using market-level data to model, identify, and estimate a dynamic discrete choice demand model for durable goods with continuous unobserved product-specific state variables. The study demonstrated that it is possible to estimate these very complicated, dynamic discrete choice models with aggregate data with just six linear regressions. In other words, they found a simple, cost-efficient way to estimate demand.
Derdenger said that when applied empirically, this method can enable companies to better understand the preferences of their consumers.
“If Nike or Samsung or Apple was able to get their hands on this aggregated industry-level sales data, they could start to understand what consumer preferences are for their goods, and then be able to forecast what demand might look like in the future or what demand might be for a new product,” he said.