US ELECTION PREDICTION USING ONLINE ADVERTISING DATA
Tuesday June 20 12:00pm
Many of us may have heard of high frequency trading, but most of us may not realize that a large proportion of the ads that we see online are bought at real time as well. Demand side platforms such as The Trade Desk use different data sources to target potential users, all done in less than 0.1 second from the moment when a user opens a browser page.
Attempting to predict the outcome of the 2016 US Election Sean D’Arcy from The Trade Desk developed Bayesian model on the basis of high frequency advertising data. In this presentation his colleague Stella will discuss what kind of data are collected for online advertising, how they can be used to find potential customers with look-a-like models and how this was successfully applied to the US election.
Stella Xu is a trading specialist at The Trade Desk a company working to change the way digital advertising is bought. She graduated with a degree in Mathematics from Australian National University and is currently pursuing a Master in Mathematics at University of New South Wales. She is passionate about increasing the efficiency of media buying and delivering the right information to the right population with data. Sean D’Arcy is the Head of Trading Innovation at The Trade Desk. His work in advertising modelling has been featured on CBS 60 minutes and has won a Young Lions Award at the Cannes Innovation Festival.