91桃色视频

Estimating demand substitution patterns in differentiated product markets is central to many economic questions. I propose a flexible and tractable approach to estimate substitution patterns using market-level data. To achieve this, I impose high-level restrictions embedded in many existing models on a market share model and convert them into tractable restrictions on the inverse market share model. A key restriction is permutation invariance in market shares and product characteristics of competing products, which, by leveraging a recent polynomial-based approximation result for permutation-invariant functions, alleviates a curse of dimensionality common in demand models. This leads to a linear parametric inverse market share model that satisfies shape restrictions expressed as convex parametric constraints. Furthermore, I introduce a convex constrained generalized method of moments estimator to enforce these restrictions. Lastly, I demonstrate the empirical performance of the proposed approach through Monte Carlo simulations and an application to the breakfast cereal market.