Author: Gordon Ji
Degree: Ph.D. in Economics
Advisor: Dr. Eugenio Miravete
Institution: University of Texas at Austin
Year: 2022
Award: Outstanding 2nd Year Paper Award

Many economic research studies have been focusing on the demand and welfare estimation of the ride-hailing market, specifically for platforms like Uber and Lyft. In this paper, I estimate the welfare effect of UberPool as a new product in the ride-hailing market, accounting for heterogeneous preferences within and across locations by using a discrete-type random coefficient nested logit model. I find that, relative to the counterfactual worlds without UberPool, UberPool can increase consumer surplus by 31.58% - 33.51%. Even a partially accessible UberPool by location is 2.57% higher on consumer surplus, compared to if only UberX were provided but with lower prices, which shows the magnitude of the variety effect in the ride-hailing market.

├── estimation/                 # Code related to demand estimation
├── prelim_graph.R              # Plots for preliminary analysis
├── supply_side.R               # Code expanding to supply side analysis
├── Counterfactual.R            # Code conducting counterfactual analysis
└── README.md                   # Project overview (this file)

This repository accompanies my research on the variety effect of UberPool. It includes:

• A discrete-type random coefficient nested logit model to estimate the demand (also known as BCS)
• Counterfactual simulations of removing UberPool

The data comes from public ride-hailing data from the City of Chicago. Large data files are ommitted. https://data.cityofchicago.org/Transportation/Transportation-Network-Providers-Trips-2018-2022-/m6dm-c72p/about_data

This project is licensed under the MIT License – see the LICENSE file for details.

If you have questions or want to collaborate, feel free to reach out:
📧 [[email protected]], [[email protected]]