Jeremy Large & Emmet Hall-Hoffarth will present a discrete choice, random utility model and a new estimation technique for analysing consumer demand for large numbers of products. In their model each product has an associated unobservable vector of attributes from which the consumer derives utility. They allow the consumer to purchase multiple products at once in a consumption bundle.
Because the number of bundles available is massive, a new estimation technique, which is based on the practice of negative sampling in NLP, is needed to sidestep an intractable likelihood function. They prove consistency of our estimator, validate the consistency result through simulation exercises, and estimate their model using supermarket scanner data.
This talk is organised by INET Oxford
This talk is live in-person at the Manor Road Building and online
To register and for more information - https://www.inet.ox.ac.uk/events/estimating-very-large-demand-systems/
Jeremy Large is a financial economist in the Department of Economics at the University of Oxford. He is also an experienced algorithmic trader on global capital markets, with a track record in Global Macro, Foreign Exchange, Listed Equity and Commodity trading. Jeremy has published research in the areas of Market Microstructure and Financial Econometrics. He lectures on these topics at graduate level, as well as in the area of Big Data and Machine Learning for Economics.
Jeremy held a Fellowship at All Souls College, Oxford from 2005 until 2008, when he joined the hedge fund AHL within Man Group Plc. In 2013 Jeremy joined the hedge fund, Tudor Investment Corporation, where he was a Quantitative Portfolio Manager. Jeremy is an investor in social enterprise and participates in the activities of the organisation, Ashoka, as a Member of the Ashoka Support Network.
Emmet Hall-Hoffarth is a DPhil candidate in the Department of Economics at the University of Oxford. His research focuses on applying machine-learning techniques to economic modelling. In particular, he has written on the use of causal-discovery algorithms to identify the structure of DSGE models and is currently working on developing estimation strategies for macroeconomic models using deep-learning. Furthermore, he works as a research assistant to Jeremy Large with whom he is involved in various projects such as the RUBE project which is the topic of this presentation.
Emmet graduated as valedictorian with a BA in economics from the University of British Columbia in 2019 and with an MPhil in economics with distinction from the University of Oxford in 2021.