CEAI Lunch and Learn: Wei Long, Patrick Button, and Chenxi Li
Please join the Center for Community-Engaged Artificial Intelligence (CEAI) and Connolly Alexander Institute for Data Science (CAIDS) for the next "Lunch and Learn" seminar exploring the impacts of data and AI on society. This session will feature two presentations: One by Economics professor Wei Long, and another by CAIDS Executive Director and Economics professor Patrick Button with Economics PhD student Chenxi Li.
Presenter: Wei Long, Associate Professor of Economics
Title: A Random Forests-Based Panel Data Approach for Program Evaluation (coauthored with Guannan Liu, Xuehong Luo, both from Xiamen University)
Abstract: It is challenging to conduct controlled experiments to assess the impacts of social policy. To address this, Hsiao et al. (2012) propose a panel data approach using factor models to estimate average treatment effects. The selection of control units is a critical step to balance the goodness of fit within-sample with the post-treatment forecasting error when the number of observed potential control units is large. In this study, we propose using random forests, an ensemble learning method, which offers robustness and requires fewer candidate models compared to existing methods. We demonstrate that our approach effectively selects almost all relevant control units, and we provide asymptotic normality results under the null of no average treatment effect and significance tests for policy interventions. Extensive simulations confirm the method's superior performance. In the empirical studies, we showcase the usefulness of the method by evaluating the impact of Brexit on the United Kingdom's GDP growth and China's anti-corruption campaign on the importation of luxury watches.
Presenters: Patrick Button, Executive Director, Connolly Alexander Institute for Data Science & Associate Professor, Department of Economics with Chenxi Li, Ph.D. Student, Department of Economics
Title: Does Racial Concordance Reduce Discrimination in Access to Mental Health Care? Preliminary Evidence from an Audit Correspondence Field Experiment
Abstract: This project is related to Simone Skeen’s project, which she presented at this seminar series recently. In this project, we seek to use the same data from an audit correspondence field experiment, described below, to determine if racial concordance reduces discrimination in access to mental health care appointments. Racial concordance refers to when healthcare providers have matching or similar racial or ethnic backgrounds to their clients or patients (e.g., Black therapists supporting their Black clients). Previous research documents that this can reduce barriers in access to healthcare and can improve health outcomes. We use data from our audit correspondence field experiment to test if Black and Hispanic therapists are less likely (compared to other therapists) to discriminate against Black and Hispanic prospective clients (compared to White prospective clients) in offering appointments. In this experiment, we send therapists appointment inquiry emails from fictitious, but realistic, prospective clients who are inquiring about appointment availability to discuss concerns relating to depression or anxiety. In the experiment, we vary the name of the prospective client to signal that they are likely Black, Hispanic, or White (e.g., Darius vs. Alejandro vs. Brian), with all other aspects held constant, just like in a randomized control trial. This allows us to measure if therapists discriminate based on race or ethnicity in how they respond to appointment inquiries. To infer the race or ethnicity of the therapist, we use AI algorithms that use public data sources to determine, with noise, the race or ethnicity of individuals based on their first and last names. We will present preliminary results of our analysis and will summarize our next steps.
Tulane University is committed to providing universal access to all our events. Please contact Meg Keenan at mkeenan@tulane.edu or 504-862-8381 for accessibility accommodations.
