Among all parameters estimated in predictive demand modeling, price elasticity — the sensitivity of demand to price changes — is by far the most difficult to measure accurately.
In markdown optimization, errors in elasticity estimation can directly translate into revenue loss or inventory misallocation, making it both a statistical and a strategic challenge.
This article explores why elasticity estimation is so difficult in real retail environments, and how modern modeling frameworks attempt to mitigate these issues.
At the SKU level, sales data is often sparse and volatile. An individual product in a specific store or segment might sell only a handful of units per week, producing an extremely low signal-to-noise ratio. When weekly demand varies more due to randomness than price, elasticity estimates become statistically meaningless.
About the author
Cyril Noirot
Lead Data Scientist
Freelance data scientist. I design and ship decision systems — forecasting, pricing, marketing measurement, optimization.