Demand Estimation with Text and Image Data

30 Pages Posted: 23 Oct 2023

See all articles by Giovanni Compiani

Giovanni Compiani

University of Chicago Booth School of Business

Ilya Morozov

Northwestern University

Stephan Seiler

Imperial College Business School; Centre for Economic Policy Research

Multiple version iconThere are 2 versions of this paper

Date Written: 2023

Abstract

We propose a demand estimation method that allows researchers to estimate substitution patterns from unstructured image and text data. We first employ a series of machine learning models to measure product similarity from products’ images and textual descriptions. We then estimate a nested logit model with product-pair specific nesting parameters that depend on the image and text similarities between products. Our framework does not require collecting product attributes for each category and can capture product similarity along dimensions that are hard to account for with observed attributes. We apply our method to a dataset describing the behavior of Amazon shoppers across several categories and show that incorporating texts and images in demand estimation helps us recover a flexible cross-price elasticity matrix.

Keywords: demand estimation, unstructured data, computer vision, text models

JEL Classification: C100, C500, C810

Suggested Citation

Compiani, Giovanni and Morozov, Ilya and Seiler, Stephan, Demand Estimation with Text and Image Data (2023). CESifo Working Paper No. 10695, Available at SSRN: https://ssrn.com/abstract=4608817 or http://dx.doi.org/10.2139/ssrn.4608817

Giovanni Compiani (Contact Author)

University of Chicago Booth School of Business ( email )

Chicago, IL
United States

Ilya Morozov

Northwestern University ( email )

2001 Sheridan Road
Evanston, IL 60208
United States

Stephan Seiler

Imperial College Business School ( email )

South Kensington Campus
Exhibition Road
London SW7 2AZ, SW7 2AZ
United Kingdom

Centre for Economic Policy Research ( email )

London
United Kingdom

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
136
Abstract Views
360
Rank
26,975
PlumX Metrics