Research

Abstract:  Understanding peer influences is essential in an increasingly interconnected world.  However, to what extent can we use data on social connections if they are incomplete?  This paper investigates peer effects in discrete choice models with incomplete data on social links.  Following Graham (2017), we set up an undirected dyadic link formation model where connections are based on homophily (similarities in characteristics) and individual fixed effects.  We identify homophily effects through available configurations among tetrads (groups of four agents).  We then identify the distribution of fixed effects through available configurations among triads (groups of three agents).  After recovering the network-generating process, we propose a simulated network approach to study the influence of peers on individual decision-making.  Simulations illustrate that the finite sample performance of the estimator is close to that obtained when the true network is observed. We apply our estimator to examine household microfinance participation decisions in rural India  (Banerjee et al., 2013), detecting positive peer effects even in cases of missing links and missing networks. 

Presentations: Asian Meeting of the Econometric Society 2024 (scheduled),  North American Summer Meeting of the Econometric Society 2024 (scheduled), TEXPOP 2024, Texas Camp Econometrics XXVII 2024, Southern Economic Association 93rd Annual Meeting 2023, Causal Data Science Meeting 2023, Seminar at Western Illinois University - Department of Economics and Decision Sciences,  18th Annual Economics Graduate Student Conference of Washington University in St. Louis,  33rd Annual Midwest Econometrics Group Conference,  SMU Brown Bag Seminar, 2nd Annual SMU Research Computing Day 2023 (Poster)

Award:  Dean's Dissertation Fellowship 2024, SMU

Abstract:  Estimating treatment effects is one of the most fundamental problems in econometric studies.  However, estimation is complicated when spillovers across individuals are present.  In this research, we focus on the spillovers in the treatment assignments and study the resulting Bayesian Nash equilibrium of a binary participation game, also known as peer-influenced propensity scores (Jackson, Lin and Yu, 2022). In addition, we allow for the case when the network is unobservable. Under network sparsity, the estimation problem is formulated as a penalized maximum likelihood estimation, where the missing binary links are treated as high-dimensional nuisance parameters. Simulation studies demonstrate the performance of our treatment effects estimators in finite samples: our approach has good performance for small networks.

Presentations: American Economic Association Annual Meeting 2024 (Poster),  1st CIREQ Interdisciplinary PhD Student Conference on Big Data and Artificial Intelligence at McGill University,  Midwest Economics Association 87th Annual Meetings 2023, Texas Camp Econometrics XXVI,  3rd Annual SMU Economics PhD Alumni Conference, Southern Economic Association 92nd Annual Meeting 2022, SMU Research & Innovation Week Graduate Poster Session 2022, SMU Microeconomics Workshop 2021

Awards:  Cobb Fellowship (Best Third-Year Paper Prize) [link]; Dean's Award, SMU Research & Innovation Week Graduate Poster Session 2022 [link]

Abstract: We study the identification and estimation of social interactions in large endogenous networks. Our analysis focuses on binary-action games in which an agent's payoff relies on one's own characteristics, the local average of beliefs about peers' actions, and a random preference shock. Endogeneity in networks results from unobservable individual characteristics that affect both link formation and the payoff. Under conditions ensuring the unique equilibrium, we establish the identification via additive separability in the payoff. The identification result is employed to develop sieve maximum likelihood estimators.

Abstract: Mental health is increasingly recognized as a public health crisis. However, mental illness is more likely than health issues to go untreated. This is particularly true in the case of adolescent mental health, where there has been growing concern over the prevalence of depression and anxiety among teenage students. One potential barrier to adolescents receiving mental health care is stigmatization from peers. Normalizing the treatment of mental health care through counseling among children can have important consequences. We assess the role of peer networks on selection into mental health care, using data from the Add Health. We do this by introducing mental health care selection, allowing for stigma, into a network model for child cognitive and social capital development. We then estimate the child’s labor market returns from receiving mental healthcare as an adolescent, given their cognitive and social capital. We find a strong financial return to counseling in adolescence which diminishes as a child’s initial social or cognitive capital grows, or if their mental health worsens.

Presentations:  12th Annual Conference of American Society of Health Economics (Poster),  Stata Texas Empirical Microeconomics Conference 2023 (Poster), SMU Microeconomics Workshop 2023, SMU Research & Innovation Week Graduate Poster Session 2023 (presented by coauthor), Southern Economic Association 92nd Annual Meeting 2022 (presented by coauthor)

Award:  Dean's Award, SMU Research & Innovation Week Graduate Poster Session 2023 (won by coauthor) [link]

Abstract:  Left-behind children are those who live with at most one parent, and receive poor economic and mental support. In this research, we study the network formation and peer effects in academic achievement for left-behind children in China. We use a first-hand dataset collected in three counties in China with over 1,600 students, where comprehensive measures of academic achievement and parental care as well as network data are available. Preliminary analyses indicate that left-behind children are more likely to make friends with those who have similar backgrounds, and that they have worse academic outcomes and mental health such as depression. We follow Bramoullé, Djebbari and Fortin (2009) to study the peer effects, and examine the relationship between parental care and the formation of social network.

Abstract:  Social network data collection often emphasizes missing data on networks rather than characteristics, leading to situations where we have abundant social network data, but missing covariates can also cause estimation bias. This paper addresses the question by formulating it as a high-dimensional estimation problem. We use machine learning techniques to predict the conditional distribution of missing covariates and recover the peer effects in a partially linear model. 

Abstract:  This study investigates the interplay among soft power, economic development, and the spatial ramifications of China’s loans to African countries. Exploring data from the China’s Overseas Development Finance (CODF) Database and the Chinese Loans to Africa (CLA) Database, covering the period from 2002 to 2019, and considering the geolocation information of African nations, we employ the synthetic control method with spillover effects to assess these dynamics. Our analysis provides a quantitative estimation of the impact of China’s loans on African nations’ soft power dynamics, encompassing cultural, religious, and political aspects, and their economic development while also considering their geographic distances. By examining the spatial spillover effects of these loans, our research contributes to a deeper understanding of the relationships between China’s economic outreach and regional economic growth in Africa.

Abstract:  Climate change poses a critical threat to global ecosystems and human societies. As the world grapples with the ongoing challenges of climate variability and extreme events, it is essential to comprehend their multifaceted impacts. In this study, we analyze the relationship between climate phenomena, amenities, and mortality, using the El Nio and Southern Oscillation (ENSO) as a natural experiment to measure the effects of climate changes on human welfare. Using meteorological data, socioeconomic factors, and epidemiological insights, we uncover how ENSO events affect mortality rates and access to essential amenities. To understand the complex dynamic between climate oscillations, infrastructure development, and human survival, we examine multiple sectors, ranging from agriculture to public health. Our findings reveal the consequences of ENSO on mortality patterns, with implications for both short-term public health interventions and long-term climate adaptation strategies.

Abstract:  Cybersecurity has become increasingly important as people rely heavily on the internet. Despite the increasing demand, there has been a severe shortage of cyber skills in the labor market. This research aims to look closely at the labor supply of cyber-trained graduates in Texas. Using the Texas Education Research Center (ERC) data, we can trace the complete education path and post-graduation labor market performance of students. We evaluate cybersecurity’s labor market efficiency by investigating Texas students’ application information, enrollment data, major decision, transfer decision, graduation dates, degree progress, and career information. The ERC data allows us to quantify the incentives and advantages to educational institutions of providing high-quality cybersecurity training. In conclusion, this research project would provide policymakers with a broader understanding of possible causes of labor supply shortages, future market trends, and policy orientation of the cybersecurity industry.

Abstract:  Compared to other powers of its time, the Ming dynasty (1368-1644) allocated a larger proportion of its resources to its military. The soldiers were responsible for growing crops and supplying the army, and Ming garrisons (Wei-suo) aimed to achieve self-sufficiency. Our analysis, based on the overlapping generations (OLG) model, indicates that the economy's steady state growth rate is zero. The question arises: what are the long-term effects of the Ming military institution? Our empirical results suggest that Ming garrisons have a lasting impact on economic development and gender equity. By studying nightlight luminosity, we find that China's historical military institution plays a significant role in present-day economic development. Moreover, the Ming garrisons had an adverse effect on the gender ratio (i.e., the balance of males and females) in 2000 and 2010. These results remain robust even after controlling for county- and province-level fixed effects, distances to post stations and garrisons, and agricultural productivity index. We also employed propensity score matching to confirm our findings.

Presentation:  SMU Macroeconomics Workshop 2022