Abstract: Online peer-to-peer marketplaces are an increasingly common venue for many types of transactions, including short-term accommodations (Airbnb), transportation (Uber), and pet-related services (Rover). Consequently, it is vital for marketplace designers to measure the impact of innovative marketplace features through randomized experimentation. In this study, we investigate new methods for the design and analysis of experiments in marketplaces, and use these techniques to measure the impact of modifying an online marketplace’s algorithmic pricing suggestions to sellers.
Speaker Bio: David Holtz is a 4th year doctoral candidate in the Information Technologies group at Sloan School of Management, MIT. His research interests span online marketplace design, causal inference, applied machine learning, and network science. David’s work thus far has focused on ratings and reviews, as well as the viability of reputation systems that don’t depend on user generated feedback. He holds an MA in Physics & Astronomy from Johns Hopkins University and a BA in Physics from Princeton University. Prior to beginning his PhD, David was a data scientist (most recently at Airbnb).