“Nonparametric Causal Estimators for Multivariate Missing Data: An Application to Estimate Treatment Effects from Digital Trace Data,” Jacqueline Mauro, UC Berkeley
QUANTITATIVE RESEARCH METHODS WORKSHOP
Abstract: In order to study the impacts on a multifaceted outcome like welfare, we extend nonparametric multivariate scaled ATE estimators to incorporate missing data, and develop a similar new LATE estimator for the IV case. We also extend the multiplier bootstrap to handle multiple outcomes at repeated measures for valid confidence bands. These estimators are doubly-robust and fully nonparametric, and allow us to robustly test for impacts of treatment on multiple outcomes at once. We apply these tools to learn the impact of short-term, high interest mobile loans in East Africa on welfare, using data from a lender in the region to develop proxy measures of welfare. The welfare impacts of these high-interest, unsecured loans is not well known; ideally they provide a new avenue to credit to the underbanked, but they may also lead to debt spirals.
Jackie Mauro is a postdoctoral scholar at the University of California - Berkeley School of Information under Joshua Blumenstock. She is studying the welfare effects of mobile banking using new nonparametric tools. Her research generally focuses on developing nonparametric causal methods motivated by real-world policy issues. These methods lean on developments in Machine Learning to create flexible yet robust estimates of causal effects. The goal is to provide practitioners across a variety of fields with the most robust possible estimates of the impacts of proposed policy changes.
This workshop series is being sponsored by the ISPS Center for the Study of American Politics and The Whitney and Betty MacMillan Center for International and Area Studies at Yale with support from the Edward J. and Dorothy Clarke Kempf Fund.
The workshop is open to Yale faculty, students, and professional staff only.