Washington University Medical Center
Washington University St. Louis: Open ScholarshipNot a member yet
26344 research outputs found
Sort by
Essays in Matching Markets
No full textMatching markets, unlike commodity markets, involve transactions where participants not only choose but also need to be chosen by others. Multiple mechanisms have been designed to address matching problems. However, their practical implementation often diverges from the theoretical ideal due to various constraints. This dissertation explores inefficiencies in centralized markets due to strategic advantages given to certain participants, frictions that constrain participants\u27 matches, and limitations on how preferences can be reported. It also proposes policies or interventions to improve market welfare and efficiency. The first chapter examines the distortions introduced by heterogeneous outside options and test scores in school choice assignments, particularly when students have incentives to misreport their preferences to a planner. The second chapter investigates two-sided matching markets in which participants can rank only a subset of potential partners, exploring the consequences of this restriction on market stability. In the third chapter, I analyze the efficiency consequences of a practical implementation of the serial dictatorship mechanism with manual assignment interventions. In the first chapter, I study the centralized allocation of students into majors under a constrained Boston mechanism, where priorities are determined by a placement test taken after students submit their preference reports to the clearinghouse. I find that, conditional on test scores, high socioeconomic status students are 50\% more likely to top-rank highly selective majors. Additionally, wealthier students are 20\% more likely to top-rank their most preferred major compared to their less-affluent counterparts. While various sources of heterogeneity between high- and low-socioeconomic students can explain these patterns, I show that if both groups had access to similar outside options, low- and high-socioeconomic students would be equally likely to top-rank their most preferred major. Moreover, if their test scores were similar, both groups would be equally likely to top-rank highly selective programs. These findings suggest that outside options and test scores provide strategic advantages to wealthier students. Therefore, I argue that policies aimed at subsidizing test scores or improving outside options for lower-income students could help reduce these strategic advantages. In the second chapter, I study the U.S. residency market. In this setting, doctors and hospitals interview with a subset of potential partners and submit preference lists to a clearinghouse, which then finds a stable matching based on those submissions. However, since agents only rank potential partners with whom they have interviewed, the resulting matching may not be stable with respect to their true preferences. This chapter investigates wasteful interviews—interviews scheduled that will not lead to a match between the participants involved. I propose an intervention by the clearinghouse during the interview stage to help participants identify these wasteful interviews. I theoretically show that, in small markets where hospitals have identical preferences, this intervention improves stability for most possible preference profiles. Additionally, I use simulations to provide evidence of this improvement in larger markets. The third chapter, co-authored with George-Levi Gayle, Peter-John Gordon, and Nadine McCloud-Rose, evaluates a modified version of the serial dictatorship mechanism implemented in Jamaica to allocate students to secondary schools nationwide. In this mechanism, students submit a constrained ranked list of schools and then take a placement test that determines the order in which they will be assigned. Each student is assigned to the highest-ranked school available on their list. However, if all the schools they ranked are at capacity, an administrator manually assigns the student to a school with an available vacancy. In this setting, some efficiency loss arises from the fact that students do not know their priorities and from the constraint on their ranked lists. We evaluate whether the administrator’s role in the system restores the efficiency lost due to constraining the ranking list length. However, we find that it does not. Instead, the administrator\u27s intervention leads to the underuse of seats at more desirable schools, compared to a scenario where an unconstrained serial dictatorship mechanism is implemented. We also find that the current mechanism is less meritocratic than the latter, in the sense that high performers are often allocated to less-preferred schools
Essays on Social Media Algorithms and Causal Inference Methods
No full textThis dissertation examines the interaction between personalized recommender systems and social media users\u27 digital consumption and privacy protection through large-scale field experiments. Motivated by the causal identification challenges commonly faced in field experiments, the dissertation further proposes a new estimator for encouragement designs when the exclusion restriction is violated. The first chapter studies how modifying a personalized recommender system to promote content diversity affects both the amount and diversity of users’ digital consumption. A fundamental trade-off in designing personalized recommender systems lies between exploitation and exploration: deciding whether to recommend familiar content that users favor or to introduce new, diverse content that may interest them in the future. We empirically examined this trade-off in a real-world-scale recommender system through a partnership with a leading global music-streaming service platform. We conducted a large-scale field experiment where users were randomly assigned to receive video recommendations from either the platform’s standard algorithm or a modified version that recommends more diverse content. Contrary to industry expectations, increasing the diversity of the recommender algorithm does not enhance users’ consumption diversity; instead, it marginally reduces their click days on the platform. However, among active users—who account for most of the platform’s content usage—a 1% increase in recommendation diversity resulted in a 0.55% increase in their consumption diversity, without affecting overall consumption levels. The increase in consumption diversity corresponds with the more accurate prediction of their consumption preferences. The results suggest that the platform should tailor its algorithm to recommend more diverse content for active users. The second chapter investigates how privacy protection measures offered by social media platforms affect users\u27 digital consumption, both intentionally and unintentionally. In this digital age, privacy concerns are escalating with the increased collection and use of personal information. Consequently, regulators have increasingly pushed companies to make the use of personal information more transparent and protect users with more privacy protection measures. However, the impacts of these policies on user behaviors and welfare remain unclear. We investigate this issue through a large-scale field experiment on a leading global social media platform. In the experiment, treated users were offered a privacy protection option to disable the People You May Know (PYMK) recommender algorithm, which could display their content to users whom the algorithm predicts are their friends elsewhere. Control users were neither informed about nor allowed to disable this function. Interestingly, we found that treated users, on average, decreased their video usage time by 0.78% compared with control users. However, the usage time of those treated users who chose to disable the function increased by 14.62% compared with a matched sample. We interpret these results as the privacy protection option having two consequences: On one hand, it raises users\u27 concerns by reminding them that their personal information is being used, thereby unintentionally reducing their usage time. On the other hand, it allows users to disable the use of personal information, which eliminates such concerns and leads to an increase in usage time as intended. To evaluate the social welfare impact of our and alternative privacy protection measures, we estimate a structural model that describes users’ decisions regarding usage and disabling the PYMK function, and use the results to run counterfactuals. The results demonstrate that different policies could lead to drastically different social welfare outcomes, highlighting the importance of considering both intended and unintended consequences. In particular, we find that lowering the costs of adopting the PYMK protection option is crucial not only for increasing overall social welfare but also for aligning the incentives of consumers and the platform, potentially creating a win-win scenario for both. The third chapter proposes a new estimator for causal inference in field experiments when the exclusion restriction is violated. Encouragement design is widely used in field experiments (or randomized controlled trials) when noncompliance in the treatment group, control group, or both is non-negligible. The standard identification strategy is to use the randomized group assignment as an instrumental variable to estimate the local average treatment effect (LATE). In many experiments, however, this instrument may violate the exclusion restriction condition, because the encouragement can directly impact the interested outcome variable. We develop a new root-n-consistent estimator using the randomized group assignment to construct an instrument that relies on the heteroskedasticity of treatment intensities between groups. Our identification strategy can recover not only LATE but also the direct impact of the encouragement on outcomes. We further propose a min-max estimator for consistent nonparametric estimation of heterogeneous treatment effects. Finally, we conducted a large-scale field experiment with a social media platform to study how expanding users\u27 social networks influences their platform usage. While ordinary least squares and standard two-stage least squares estimators report a positive effect, our estimator suggests that the effect comes solely from the encouragement. We find evidence supporting the null effect of network expansion, indicating that firms may waste resources on false positives when the exclusion restriction is violated in their field experiments
Supporting Employed Informal Caregivers
Get PDFMore than 53 million Americans are informal caregivers. These family members, friends, and acquaintances provide social, medical, transportation, and other support to individuals because of illness or disability without getting paid. Additionally, at least 2 million grandparents, not counting other kinship caregivers, provide care for children who cannot remain with their parents.
Functional Equivariance and Backward Error Analysis
Get PDFGeometric, or structure-preserving, numerical integration has long been used as a framework for studying integrators that preserve a systems invariants. In backward error analysis, the approximate numerical flow of a system is viewed as the exact flow of a modified problem, allowing us to gain qualitative insights into the behavior of the numerical solution. The preservation of conservation laws by a numerical integrator can be generalized to -functionally equivariant integrators, where represents an observable of the system in consideration. This thesis describes the behavior of geometric integrators through the lens of backward error analysis. First, we extend the idea of -functional equivariance to modified vector fields, generalizing results on invariant preservation and describing the numerical evolution of non-invariant observables. Next, we introduce algebraic conditions for -functionally equivariant B-series methods and their modified vector fields. A special case of this condition, when is quadratic, yields the well-known algebraic characterization of symplectic B-series. Finally, we define conjugate functionally equivariant integrators and develop the notion of modified observables with respect to such integrators, analogous to modified quadratic observables, in the context of near symplectic integrators
Understanding the Effects of Genetic Variation on Maize and Pennycress Root System Architectures and Their Interaction with Nitrogen
Get PDFRoots are a critical yet hidden part of plants, providing physical support and access to resources such as nutrients and water. The development of a plant’s root system across time and space is referred to as root system architecture, which influences the effectiveness of resource capture. In modern agriculture, applications of nitrogen fertilizers are critical to maintain high yield. However, up to 70% of applied nitrogen is not utilized by the plant and escapes into waterways and the atmosphere, causing environmental harm. Efficient nitrogen uptake is limited by the ability of the plant to access and absorb nitrogen through its roots. It has been proposed that modifying root system architecture could improve access to and uptake of nitrogen from the soil, for example, by modifying the postembryonic shoot borne root system comprised of nodal and lateral roots to increase soil exploration and grow deeper roots. But to test these theories and select for desirable root traits, a better understanding of the genetic basis of root system architecture is needed. In this dissertation, we identify a core regulator of nodal rooting from the classic maize mutant Rootless, which we name, Rootless1 (ZmRt1). We demonstrate that ZmRt1 quantitatively regulates nodal root primordia development across several maize backgrounds. We identified a Ds insertion allele, designated rt1-2, which results in a nodal rooting phenotype that we hypothesized would result in a root system architecture that improves nitrogen uptake in the field. To test this, we conducted nitrogen contrast experiments where we used advanced root phenomics tools to show changes in the root system architecture of rt1-2 plants leading to improved nitrogen uptake and yield. Additionally, we conducted the first study to measure the root system architecture diversity in the emerging cover crop pennycress. Adopting pennycress as a cover crop has the potential to improve the sustainability of the maize agroecosystem by catching residual nitrogen in fields. Together, this work advanced our understanding of the genetic regulation of RSA through ZmRt1 and natural variation in pennycress and maize RSA in the field, providing a valuable resource for future genetic studies
Diverse Cell Types Establish a Pathogenic Immune Environment in Peripheral Neuropathy
Get PDFNeuroinflammation plays a complex and context-dependent role in many neurodegenerative diseases. We identified a key pathogenic function of macrophages in a mouse model of a rare human congenital neuropathy in which SARM1, the central executioner of axon degeneration, is activated by hypomorphic mutations in the axon survival factor NMNAT2. Macrophage depletion blocked and reversed neuropathic phenotypes in this sarmopathy model, revealing SARM1-dependent neuroimmune mechanisms as key drivers of disease pathogenesis. In this study, we investigated the impact of chronic subacute SARM1 activation on the peripheral nerve milieu using single cell/nucleus RNA-sequencing (sc/snRNA-seq). Our analyses reveal an expansion of immune cells (macrophages and T lymphocytes) and repair Schwann cells, as well as significant transcriptional alterations to a wide range of nerve-resident cell types. Notably, endoneurial fibroblasts show increased expression of chemokines (Ccl9, Cxcl5) and complement components (C3, C4b, C6) in response to chronic SARM1 activation, indicating enhanced immune cell recruitment and immune response regulation by non-immune nerve-resident cells. Analysis of CD45+ immune cells in sciatic nerves revealed an expansion of an Il1b+ macrophage subpopulation with increased expression of markers associated with phagocytosis and T cell activation/proliferation. We also found a significant increase in T cells in sarmopathic nerves. Remarkably, T cell depletion rescued motor phenotypes in the sarmopathy model. These findings delineate the significant changes chronic SARM1 activation induces in peripheral nerves and highlights the potential of immunomodulatory therapies for SARM1-dependent peripheral neurodegenerative disease
Relationships between Household Economic Distress and Child Mental Health: Determinants, Mechanisms, and Interventions
Get PDFEconomic inequality has grown dramatically worldwide, creating serious problems for children and families. At the same time, child and adolescent mental health problems have reached crisis levels, yet many of them cannot access treatment. Researchers have studied how economic inequality affects mental and behavioral health, but most of this work focuses on adults rather than children. This dissertation used recent data from the Panel Study of Income Dynamics (2019 and 2021 waves) to unpack the complex relationships between household economic distress and child mental health across multiple levels, including individual, family, school, and community factors in the United States. This dissertation includes three interconnected components. The first part used machine learning models to identify factors that predict household income and asset poverty across different levels. The second study examined how household economic distress influences child mental health through different pathways, using structural equation models to understand these mechanisms across individual, household, school, and community levels. The third study tested whether families with child savings accounts have better child mental health outcomes, using generalized linear models combined with propensity score weighting. The results reveal important determinants and mechanisms and offer practical insights for interventions and policies that could support children and families at multiple levels
New Dynamometer Setup
Get PDFThe goal of this report is to go over the design and implementation of a dyno system for the WashU FSAE team. Utiliing existing infrastrucure and devices from 2008 and 2016 and learning how they work, a new system was designed and implemented. With cost in mind and the fact that the car is dissassembled each year, a water-brake engine dyno was designed, manufactured, and created for the Yamaha R6 engine used by the team. The main limitation of water-brake systems with motorcycle engines is the torque of motorcycle engines is too much for a water-brake due to the transmission being built into the engine: the water-brake attaches to the output of the transmission instead of to the crankshaft. Thus, a gear reduction is needed, a 0.4 gear ratio in this case, to decrease torque and increase speed at the water-brake. With this, many factors had to be added into consideration for the design such as vibrations, shaft misalignment, chain tensioning, engine ease of modification, and more as well as the water-brake support system. While there were some oversights in the design that became apparent as it was being assembled, these challenges were able to be overcome. While the dyno was not up and running before the semester was over, it is 95% complete and should be operational next semester
Out on Olive: Exploring Queer Nightlife Histories, An Exhibit at the National Building Arts Center
Get PDFFinding Queer Nightlife History is an exhibit at the National Building Arts Center that invites visitors to explore four nightlife venues connected to a series of windows salvaged from 4236-4240 Olive. Each window is a literal and metaphorical window into one of four establishments that served St. Louis\u27 Olive Street between 1930-1964. By exploring each of the four establishments, visitors discover different stories of St. Louis\u27 past and that Queer history IS history. As a result, this exhibit weaves together Queer histories, building arts, and the bittersweet process of finding oneself in an archive.
Mendel Sato Research Award 2025, Undergraduate studen
Buy-Now-Pay-Later and Alternative-Financial-Service Use Among Low-Wage Workers
Get PDFBuy-now-pay-later (BNPL) services offer consumers the option of short-term interest-free loans for retail purchases, but the rapid growth of such services has raised concerns about indebtedness and impacts on household financial security. This brief draws upon data from the nationally representative Workforce Economic Inclusion and Mobility survey to examine the use of BNPL offerings, the demographics of users, the use of alternative credit products, and differences in BNPL product use by employment status