Traditional price elasticity models often suffer from inaccurate estimations due to the influence of confounding variables.
Suppose you want to understand the impact of changing your regular/list prices, the impact of promotional discounts, the impact of merchandising support from your retail partners, and the adverse impacts of competitors making price investments.
Like most people, you build a linear regression model, perhaps including some interaction terms and some regularization. This approach still dominates most price elasticity modeling practices.
Complications arise from both confounding and omitted variables.
+ Seasonality impacts your unit sales, promotional investments, retailer...